Home

Microdata User Guide

Contents

1. 7 6 Weighting The principle behind estimation in a probability sample such as the LFS is that each person in the sample represents besides himself or herself several other persons not in the sample For example in a simple random 296 sample of the population each person in the sample represents 50 persons in the population The weighting phase is a step which calculates for each record what this number is This weight appears on the microdata file and must be used to derive meaningful estimates from the survey For example if the number of persons eligible for EI benefits is to be estimated it is done by selecting the records referring to those individuals in the sample with that characteristic and summing the weights entered on those records Details of the method used to calculate these weights are presented in Chapter 11 0 7 7 Suppression of Confidential Information It should be noted that the Public Use Microdata Files PUMF may differ from the survey master files held by Statistics Canada These differences usually are the result of actions taken to protect the anonymity of individual survey respondents The most common actions are the suppression of data items and grouping values into wider categories For certain variables that are susceptible to identifying individuals the PUMF may have been treated with local suppression that is some of the values in the master file may have been coded as not stated on the PUMF
2. The survey master file includes geographic identifiers for the 10 provinces and for the El economic regions The PUMF does not contain any geographic identifiers below the provincial level and some provinces were grouped i e Atlantic region and Manitoba with Saskatchewan Grouping of provinces was done to avoid excessive data suppression of useful variables Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide The survey master file includes the respondent s precise age while the PUMF contains age groups only Similarly detailed industry and occupation job tenure number of months since last worked age of the baby in months mothers only and several other detailed variables are only available on the survey master file Users requiring access to information excluded from the microdata files may purchase custom tabulations Estimates generated will be released to the user subject to meeting the guidelines for analysis and release outlined in Chapter 9 0 of this document Special Surveys Division 29 Employment Insurance Coverage Survey 2006 User Guide 8 0 Data Quality 8 1 Response Rates The following tables summarize the number of in scope persons number of respondents and resulting response rate to the Employment Insurance Coverage Survey EICS In scope Province Sample Response Response Rate 96 Newfoundland and Labrador 604 503 83 Prince Edward Island 346 305 88 Nova Sco
3. Microdata User Guide Employment Insurance Coverage Survey 2006 Bel cese meis Canada Employment Insurance Coverage Survey 2006 User Guide Table of Contents 1 0 Introduction oi S 5 2 0 Izd pei 7 3 0 olx cS 9 4 0 Concepts and Definitions uicti rrt ann tene tas 11 4 1 Labour Force Survey Concepts and Definitions sse 11 4 2 Employment Insurance Coverage Survey Concepts and Definitions 12 5 0 Survey Methodology 15 5 1 Population Coverage c om cce ter tpe ed te tede tested dci bees 15 5 2 Sample Design iiio itti n e v ein 15 5 2 1 Primary Stratification sessssssssssssssesesee eee en nennen nnne 15 5i2 2 TYPOS OR AGAS ureien ei ie ea aaaea er EIER RE dennen 15 5 2 3 Secondary Stratification nsun anan a a a a 16 5 2 4 Cluster Delineation and Selection sssssssse eee 16 5 2 5 Dwelling Selection teer e eti en Beine 17 5 2 6 Person Selection e Rd Ri 17 5 3 Sample ZO ico lies ea is lien 17 5 4 Samiple Rotaliol 5 du on ore tene e REO nie EE enn re 17 5 5 Modifications to the Labour Force Survey Design for the Employment Insurance Coverage UV leds e ter fas 17 5 5 1 Target Population recie iret een 18 5 5 2 Type 4 A Special Case nnersnsenssennnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnannnnnnnnnnnnn 18 55 3 SUB SAMPIING cete Dr
4. through message screens on the computer to modify the information However for some questions interviewers have the option of bypassing the edits and of skipping questions if the respondent does not know the answer or refuses to answer Therefore the response data are subjected to further edit and imputation processes once they arrive in head office 7 2 Verification and Editing Electronic text files containing the daily transmissions of completed cases are combined to create the raw survey file At the end of collection this file should contain one record for each sampled individual Before further processing verification is performed to identify and eliminate potential duplicate records and to drop non response and out of scope records There are a number of circumstances where respondents may be found out of scope of the EICS By far the majority of out of scope sampled cases are found among Type 4 respondents refer to Section 5 5 2 A small number of other records are dropped after verifying the accuracy of the information used in sampling Finally a very small percentage of the sample is no longer in scope of the EICS at time of the interview due to death moving to an institution or moving outside of the country A criterion is defined for dropping non response records In the EICS the respondent must have at least responded to the items required to derive the Employment Insurance El coverage variable COV refer to Section 7 5
5. while 775 530 1 172 069 66 2 of the regular population in Ontario contributed to El How does the user determine the coefficient of variation of the difference between these two estimates 1 Using the MOTHER 0 and TYPE 2 or 4 QUEBEC coefficient of variation table and the MOTHER 0 and TYPE 2 or 4 ONTARIO coefficient of variation table in the same manner as described in Example 2 gives the CV of the estimate for Quebec as 2 5 and the CV of the estimate for Ontario as 2 5 Special Surveys Division 47 Employment Insurance Coverage Survey 2006 User Guide Employment Insurance Coverage Survey 2000 to 2003 Approximate Sampling Variability Tables MOTHER 0 and TYPE z 2 or 4 Quebec NUMERATOR OF ESTIMATED PERCENTAGE PERCENTAGE 000 0 1 1 0 2 0 5 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 1032 1027 101 1 984 9 7 928 899 868 83 6 80 4 rr 730 726 715 69 6 676 656 635 614 59 1 568 596 593 584 568 552 536 519 501 483 464 516 514 506 492 478 464 449 434 418 40 2 462 459 452 440 428 415 402 388 374 359 421 419 413 402 391 379 367 354 341 328 srr 390 388 382 372 362 31 340 328 31 6 304 RRR KK kkk kk kkk kkk RR KEK kkk RRR 7 3 7 1 6 8 6 6 KR kkkkkk kkk kkk kkkkkk RRR IK KK RRR RK KR 1 0 5 9 5 7 KR RRR IK KK dokekekelek RIK KKK RRR RK KR KR RRR KK 5 1 RRR RRR EK KARR kkkkkk RRR KARR Klek kkk kkk KEKE Klek RRR KR RK IK KK kkkkkk RRR RK RRR KK RRR KR Kikke RK KKK kkk KK RRR KK KR RRR doekekejek eK KERR RRR KR
6. 6 557 542 527 511 495 478 46 1 442 404 icon 49 0 482 469 456 442 428 414 39 9 383 35 0 ici 438 431 420 408 396 383 37 0 357 343 31 3 ener 40 0 394 383 37 2 361 3 0 338 326 31 3 28 6 HIPS 37 0 365 355 345 334 324 313 302 29 0 26 4 KR RK IK RRR RRR dokdekelek Kekke RK 5 4 RRR KK kkk kk RAKE RRR KK keke kkk kkk 4 9 KKK RRR IK KKK dokekekelek RIK KK RRR KKK KRRKRKKK RAKE kkkkkk kkk RRR KARR 750 RRR kkk aR RK RRR RK RRR RRR Kikke RK RRR RRR KE KKK RRR IK 1 000 RRR kkkkkk RRR kiek RRR kkkk kk kkk kkkkkk kkk KARE RRR RRR RK NOTE For correct usage of these tables please refer to the microdata documentation Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 2 The estimated aggregate 400 393 does not appear in the left hand column the Numerator of Percentage column so it is necessary to use the figure closest to it namely 400 000 3 The coefficient of variation for an estimated aggregate is found by referring to the first non asterisk entry on that row namely 4 0 4 Sothe approximate coefficient of variation of the estimate is 4 096 The finding that there were 400 393 to be rounded according to the rounding guidelines in Section 9 1 unemployed individuals received regular El benefits during the reference week is publishable with no qualifications Example 2 Estimates of Proportions or Percentages of Persons Possessing a Characteristic Suppose that the user estimates that 605 7
7. calculated as MUR X 100 3 7 0 81 There is more information on the calculation of coefficients of variation in Chapter 10 0 Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 9 0 Guidelines for Tabulation Analysis and Release This chapter of the documentation outlines the guidelines to be adhered to by users tabulating analyzing publishing or otherwise releasing any data derived from the survey microdata files With the aid of these guidelines users of microdata should be able to produce the same figures as those produced by Statistics Canada and at the same time will be able to develop currently unpublished figures in a manner consistent with these established guidelines 9 1 Rounding Guidelines In order that estimates for publication or other release derived from these microdata files correspond to those produced by Statistics Canada users are urged to adhere to the following guidelines regarding the rounding of such estimates a Estimates in the main body of a statistical table are to be rounded to the nearest hundred units using the normal rounding technique In normal rounding if the first or only digit to be dropped is O to 4 the last digit to be retained is not changed If the first or only digit to be dropped is 5 to 9 the last digit to be retained is raised by one For example in normal rounding to the nearest 100 if the last two digits are between 00 and 49 they are changed t
8. metropolitan areas rotation groups household and economic family size Weights are also adjusted so that estimates of the previous month s industry and labour status estimates derived from the present month s sample sum up to the corresponding estimates from the previous month s sample This is called composite estimation The entire adjustment is applied using the generalized regression technique This final weight is normally not used in the weighting for a supplement to the LFS Instead it is the sub weight which is used as explained in the following paragraphs 11 2 Weighting Procedures for the Employment Insurance Coverage Survey The principles behind the calculation of the weights for the EICS are identical to those for the LFS However further adjustments are made to the LFS sub weights in order to derive a final weight for the individual records on the EICS microdata file 1 An adjustment to account for the use of a one sixth sub sample instead of the full LFS sample In the case of the mothers the fraction is two sixths 2 An adjustment to account for the EICS sub sampling refer to Section 5 5 1 3 An adjustment to account for the additional non response to the supplementary survey i e non response to the EICS for individuals who did respond to the LFS or for which previous month s LFS data was brought forward The procedure is similar to the LFS non response weight adjustment but groupings are based on different variabl
9. microdata file because of confidentiality Variances that take the complete sample design into account can be calculated for many statistics by Statistics Canada on a cost recovery basis The method available to approximate the true variance is to use a replication method namely the bootstrap method This method is known to correctly approximate the true value of the variance A file containing 1 000 bootstrap weights is available Variance calculation using 1 000 bootstrap weights involves calculating the estimates with each of these 1 000 weights and then calculating the variance of these 1 000 estimates 9 5 Coefficient of Variation Release Guidelines Before releasing and or publishing any estimates from the EICS users should first determine the quality level of the estimate The quality levels are acceptable marginal and unacceptable Data quality is affected by both sampling and non sampling errors as discussed in Chapter 8 0 However for this purpose the quality level of an estimate will be determined only on the basis of sampling error as reflected by the coefficient of variation as shown in the table below Nonetheless users should be sure to read Chapter 8 0 to be more fully aware of the quality characteristics of these data First the number of respondents who contribute to the calculation of the estimate should be determined If this number is less than 30 the weighted estimate should be considered to be of unacceptable quality F
10. population in Ontario who contributed to El From Example 3 Section 10 1 1 the standard error of the difference between these two estimates was found to be 0 025 Hence X X 0 757 0 662 0 005 _ 0 0 025 0 025 d t 3 80 Since t 3 80 is greater than 2 it must be concluded that there is a significant difference between the two estimates at the 0 05 level of significance 10 4 Coefficients of Variation for Quantitative Estimates For quantitative estimates special tables would have to be produced to determine their sampling error Since most of the variables for the EICS are primarily categorical in nature this has not been done As a general rule however the coefficient of variation of a quantitative total will be larger than the coefficient of variation of the corresponding category estimate i e the estimate of the number of persons contributing to the quantitative estimate If the corresponding category estimate is not releasable the quantitative estimate will not be either For example the coefficient of variation of the total number of unemployed receiving regular El benefits would be greater than the coefficient of variation of the corresponding proportion of unemployed receiving regular El benefits Hence if the coefficient of variation of the proportion is unacceptable making the proportion not releasable then the coefficient of variation of the corresponding quantitative estimate will also be unacceptable making
11. reference the birth or adoption of a child and also between working and non working respondents to make the questions more relevant to the respondent s situation In most cases the derived variable was created by simply combining answers from two questions This is the case for respondents working at the time of the interview WORKNOW type of benefit received in the reference week or month BENTYP duration of the benefits BENWEEKS benefits amount BENAMNT receiving advance notice for work interruption NOTICE taking a break from work after birth or adoption BREAKWRK and parental benefits claimed by the spouse SPCLAIM Similarly questions regarding employment after birth or adoption EMPAGREE SAMEMP and WORKCOND and on childcare arrangements CHLDCARE were asked differently for mothers on leave than for mothers who had already returned to work 7 5 3 Combining Data From the Labour Force Survey and the Employment Insurance Coverage Survey Questions related to the employer and employment conditions were only asked in the EICS if the information was not available from the Labour Force Survey LFS In the LFS these questions relate to the current job or for some items to the previous job if held in the previous year The EICS is looking for this information for all respondents who worked in the previous two years Generally the variable name used in the LFS microdata file was used FTPT HRLYEARN Many of these employment re
12. the quantitative estimate not releasable Coefficients of variation of such estimates can be derived as required for a specific estimate using a technique known as pseudo replication This involves dividing the records on the microdata files into subgroups or replicates and determining the variation in the estimate from replicate to replicate Users wishing to derive coefficients of variation for quantitative estimates may contact Statistics Canada for advice on the allocation of records to appropriate replicates and the formulae to be used in these calculations 10 5 Coefficient of Variation Tables Refer to EICS2006 CVTabsE doc for the coefficient of variation tables Special Surveys Division 53 Employment Insurance Coverage Survey 2006 User Guide 11 0 Weighting Since the Employment Insurance Coverage Survey EICS used a sub sample of the Labour Force Survey LFS sample the derivation of weights for the survey records is clearly tied to the weighting procedure used for the LFS The LFS weighting procedure is briefly described below 11 1 Weighting Procedures for the Labour Force Survey In the LFS the final weight attached to each record is the product of the following factors the basic weight the cluster sub weight the stabilization weight the balancing factor for non response and the province age sex and sub provincial area ratio adjustment factor Each is described below Basic Weight In a probability sample th
13. 18 largest cities across Canada The purpose of this is to ensure better representation of apartment dwellers in the sample as well as to minimize the effect of growth in clusters due to construction of new apartment buildings In the major cities the apartment strata are further stratified into low income strata and regular strata Where it is possible and or necessary the urban area frame is further stratified into regular strata high income strata and low population density strata Most urban areas fall into the regular urban strata which in fact cover the majority of Canada s population High income strata are found in major urban areas while low density urban strata consist of small towns that are geographically scattered In rural areas the population density can vary greatly from relatively high population density areas to low population density areas resulting in the formation of strata that reflect these variations The different stratification strategies for rural areas were based not only on concentration of population but also on cost efficiency and interviewer constraints In each province remote settlements are sampled proportional to the number of dwellings in the settlement with no further stratification taking place Dwellings are selected using systematic sampling in each of the places sampled 5 2 4 Cluster Delineation and Selection Households in final strata are not selected directly Instead each stratum is divid
14. 40 600 Ontario 184 600 amp over 54 800 to lt 184 600 under 54 800 Manitoba and Saskatchewan 36 700 8 over 11 700 to lt 36 700 under 11 700 Alberta 64 000 over 20 500 to lt 64 000 under 20 500 British Columbia 83 000 over 26 900 to lt 83 000 under 26 900 Western Provinces 93 400 8 over 26 100 to lt 93 400 under 26 100 Canada 163 800 amp over 42 800 to lt 163 800 under 42 800 Province and Region for Acceptable CV Marginal CV Unacceptable CV MOTHER 0 and TYPE 2 or 4 0 0 to 16 5 16 6 to 33 3 gt 33 3 Atlantic Provinces 64 800 amp over 20 400 to lt 64 800 under 20 400 Quebec 162 100 over 46 500 to lt 162 100 under 46 500 Ontario 205 400 amp over 56 600 to lt 205 400 under 56 600 Manitoba and Saskatchewan 43 300 amp over 12 100 to 43 300 under 12 100 Alberta 98 600 over 29 400 to lt 98 600 under 29 400 British Columbia 51 500 over 13 600 to lt 51 500 under 13 600 Western Provinces 64 000 over 16 400 to lt 64 000 under 16 400 Canada 147 600 amp over 37 300 to lt 147 600 under 37 300 Special Surveys Division 41 Employment Insurance Coverage Survey 2006 User Guide 10 0 Approximate Sampling Variability Tables In order to supply coefficients of variation CV which would be applicable to a wide variety of categorical estimates produced from this microdata file and which could be readily accessed by the user a set of Approximate Sampling Variability Tables has been p
15. 5 Editing consists in modifying the data at the individual variable level The first step in editing is to determine which items from the survey output need to be kept on the survey master file Subsequently invalid characters are deleted and the data items are formatted appropriately Text fields are stripped off the main files and written to a separate file for coding The first type of error treated was errors in questionnaire flow where questions which did not apply to the respondent and should therefore not have been answered were found to contain answers In this case a computer edit automatically eliminated superfluous data by following the flow of the questionnaire implied by answers to previous and in some cases subsequent questions For skips based on answered questions all skipped questions are set to Valid skip 6 96 996 etc For skips based on Don t know or Refusal all skipped questions are set to Not stated 9 99 999 etc The remaining empty items are filled with a numeric value 9 99 999 etc depending on variable length These codes are reserved for processing purposes and mean that the item was Not stated Special Surveys Division 23 Employment Insurance Coverage Survey 2006 User Guide 24 There was no other type of editing or imputation done on questionnaire items Therefore some internal inconsistency may become apparent when conducting analysis One notable example is the item on hou
16. 700 under 33 400 Ontario 115 900 amp over 36 300 to lt 115 900 under 36 300 Manitoba and Saskatchewan 18 600 amp over 6 700 to lt 18 600 under 6 700 Alberta 31 700 8 over 13 400 to lt 31 700 under 13 400 British Columbia 47 300 over 17 600 to lt 47 300 under 17 600 Western Provinces 54 700 8 over 16 800 to lt 54 700 under 16 800 Canada 40 400 amp over 10 200 to lt 40 400 under 10 200 Province and Region for Acceptable CV Marginal CV Unacceptable CV MOTHER 1 0 0 to 16 5 16 6 to 33 3 gt 33 3 Atlantic Provinces 6 400 amp over 2 000 to lt 6 400 under 2 000 Quebec 37 200 8 over 13 300 to 37 200 under 13 300 Ontario 38 900 amp over 11 900 to 38 900 under 11 900 Manitoba and Saskatchewan 7 800 amp over 2 500 to 7 800 under 2 500 Alberta 18 800 amp over 6 400 to 18 800 under 6 400 British Columbia 18 100 amp over 7 700 to lt 18 100 under 7 700 Western Provinces 22 900 amp over 6 700 to lt 22 900 under 6 700 Canada 38 600 amp over 10 300 to lt 38 600 under 10 300 40 Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide Province and Region for Acceptable CV Marginal CV Unacceptable CV MOTHER 0 and TYPE 3 0 0 to 16 5 16 6 to 33 3 gt 33 3 Atlantic Provinces 39 100 amp over 11 500 to lt 39 100 under 11 500 Quebec 129 100 amp over 40 600 to lt 129 100 under
17. 77 740 586 81 896 of unemployed individuals potentially eligible to receive El benefits were eligible to receive El benefits How does the user determine the coefficient of variation of this estimate 1 Refer to the coefficient of variation table for TYPE 1 CANADA 2 Because the estimate is a percentage which is based on a subset of the total population i e unemployed individuals potentially eligible to receive El benefits it is necessary to use both the percentage 81 896 and the numerator portion of the percentage 605 777 in determining the coefficient of variation 3 The numerator 605 777 does not appear in the left hand column the Numerator of Percentage column so it is necessary to use the figure closest to it namely 500 000 Similarly the percentage estimate does not appear as any of the column headings so it is necessary to use the percentage closest to it 90 0 4 The figure at the intersection of the row and column used namely 1 4 is the coefficient of variation to be used 5 Sothe approximate coefficient of variation of the estimate is 1 496 The finding that 81 8 of unemployed individuals potentially eligible to receive El benefits were eligible to receive El benefits can be published with no qualifications Example 3 Estimates of Differences Between Aggregates or Percentages Suppose that a user estimates that 543 846 718 300 75 796 of the regular employed population in Quebec contributed to El
18. Alberta 1 565 British Columbia 1 567 Canada 14 720 Special Surveys Division 19 Employment Insurance Coverage Survey 2006 User Guide 6 0 Data Collection Data collection for the Labour Force Survey LFS is carried out each month during the week following the LFS reference week The reference week is normally the week containing the 15 day ofthe month 6 1 Interviewing for the Labour Force Survey Statistics Canada interviewers are employees hired and trained to carry out the LFS and other household surveys Each month they contact the sampled dwellings to obtain the required labour force information Each interviewer contacts approximately 75 dwellings per month Dwellings new to the sample are usually contacted through a personal visit using the computer assisted personal interview CAPI The interviewer first obtains socio demographic information for each household member and then obtains labour force information for all members aged 15 and over who are not members of the regular armed forces Provided there is a telephone in the dwelling and permission has been granted subsequent interviews are conducted by telephone This is done out of a centralized computer assisted telephone interviewing CATI unit where cases are assigned randomly to interviewers As a result approximately 85 of all households are interviewed by telephone In these subsequent monthly interviews the interviewer confirms the socio demographic informa
19. Insurance Coverage Survey 2006 User Guide 5 2 5 Dwelling Selection In all three types of areas urban rural and remote areas selected clusters are first visited by enumerators in the field and a listing of all private dwellings in the cluster is prepared From the listing a sample of dwellings is then selected The sample yield depends on the type of stratum For example in the urban area frame sample yields are either six or eight dwellings depending on the size of the city In the urban apartment frame each cluster yields five dwellings while in the rural areas and EA parts of cities each cluster yields 10 dwellings In all clusters dwellings are sampled systematically This represents the final stage of sampling 5 2 6 Person Selection Demographic information is obtained for all persons in a household for whom the selected dwelling is the usual place of residence LFS information is obtained for all civilian household members 15 years of age or older Respondent burden is minimized for the elderly age 70 and over by carrying forward their responses for the initial interview to the subsequent five months in the survey 5 3 Sample Size The sample size of eligible persons in the LFS is determined so as to meet the statistical precision requirements for various labour force characteristics at the provincial and sub provincial level to meet the requirement of federal provincial and municipal governments as well as a host of oth
20. Labour Force Survey Questionnaire uunssseensnnennsnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn nen 59 12 2 The Employment Insurance Coverage Survey Questionnaires susssss 59 Record Layout with Univariate Frequencies aeree eene 61 Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 1 0 Introduction The Employment Insurance Coverage Survey EICS was conducted by Statistics Canada with the cooperation and support of Human Resources and Social Development Canada formerly Human Resources and Skills Development Canada This manual has been produced to facilitate the use of the microdata and the interpretation of the survey results Any question about the data set or its use should be directed to Statistics Canada Client Services Special Surveys Division Telephone 613 951 3321 or call toll free 1 800 461 9050 Fax 613 951 4527 E mail ssd statcan ca Special Surveys Division 5 Employment Insurance Coverage Survey 2006 User Guide 2 0 Background The Employment Insurance Coverage Survey EICS was launched in 1997 primarily in response to a need to better understand the relationship between the number of persons in receipt of Employment Insurance EI benefits and the number of unemployed as reported by the Labour Force Survey The EI administrative data is limited with respect to the population covered and the variables availabl
21. RRR kkk kk kkk kkk Kiek RRR KARR dokdekeiek RRR doekekek kkk kkkkkk NOTE For correct usage of these tables please refer to the microdata documentation Employment Insurance Coverage Survey 2000 to 2003 Approximate Sampling Variability Tables MOTHER 0 and TYPE 2 or 4 Ontario NUMERATOR OF ESTIMATED PERCENTAGE PERCENTAGE 000 0 1 1 0 2 0 5 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 50 0 122 9 122 4 121 8 119 9 116 7 113 4 110 0 106 5 102 9 99 2 95 3 87 0 mere 865 8 1 88 825 80 2 77 8 753 728 701 67 4 61 5 70 7 703 692 674 655 635 615 59 4 572 55 0 502 61 2 60 9 599 58 3 56 7 55 0 533 514 496 476 43 5 meres 547 544 536 522 507 49 2 476 460 443 426 38 9 50 0 49 7 489 476 463 449 435 420 405 389 35 5 46 3 460 453 441 429 41 6 403 389 37 5 36 0 kkkkkk kkkkkk 55 kkk kkk Joekdekeke kkk kkkkkk kkk kkk kkk kk Kekke kkk 1 000 kkkkkk kkk kkkkkk NOTE For correct usage of these tables please refer to the microdata documentation Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 2 Using Rule 3 the standard error of a difference X x is where x is estimate 1 Quebec jen is estimate 2 Ontario and and are the coefficients of variation of X and x l espectively That is the standard error of the difference d 0 757 0 662 0 095 is c 0 757 X0 025 0 662 X0 025 J 4 0 0003581 0 0002739 0 025 3 The coe
22. The beneficiary unemployed B U ratio is calculated for a given week by dividing the number of regular El beneficiaries by the total number of unemployed people In 2006 53 of the unemployed were potentially eligible for Employment Insurance Of those who were potentially eligible 83 could Special Surveys Division 9 Employment Insurance Coverage Survey 2006 User Guide meet the entrance requirements of the program and were very likely to receive benefits during their unemployment spell if they claimed The remaining 1796 did not have enough 10 Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 40 Concepts and Definitions This chapter outlines concepts and definitions of interest to the users The concepts and definitions used in the Labour Force Survey LFS are described in Section 4 1 while those specific to the Employment Insurance Coverage Survey EICS are given in Section 4 2 Users are referred to Chapter 12 0 of this document for a copy of the actual survey questionnaire s used 4 1 Labour Force Survey Concepts and Definitions Labour Force Status Designates the status of the respondent vis vis the labour market a member of the non institutional population 15 years of age and over is either employed unemployed or not in the labour force Employment Employed persons are those who during the reference week a did any work at all at a job or business or b hada job but were
23. ain a weighted average of the form X Y the numerator x is calculated as for a quantitative estimate and the denominator Y is calculated as for a categorical estimate For example to estimate the average number of weeks EI was received by mothers a estimate the total number of weeks X as described above b estimate the number of mothers currently working Y in this category by summing the final weights of all records with MOTHER 1 and WORKNOW 1 then C divide estimate a by estimate b X IY 9 4 Guidelines for Statistical Analysis The EICS is based upon a complex sample design with stratification multiple stages of selection and unequal probabilities of selection of respondents Using data from such complex surveys presents problems to analysts because the survey design and the selection probabilities affect the estimation and variance calculation procedures that should be used In order for survey estimates and analyses to be free from bias the survey weights must be used While many analysis procedures found in statistical packages allow weights to be used the meaning or definition of the weight in these procedures may differ from that which is appropriate in a sample survey framework with the result that while in many cases the estimates produced by the packages are correct the variances that are calculated are poor Approximate variances for simple estimates such as totals proportions and rat
24. al estimate could become acceptable based on the exact CV calculation Remember f the number of observations on which an estimate is based is less than 30 the weighted 44 estimate is most likely unacceptable and Statistics Canada recommends not to release such an estimate regardless of the value of the coefficient of variation 10 1 How to Use the Coefficient of Variation Tables for Categorical Estimates The following rules should enable the user to determine the approximate coefficients of variation from the Approximate Sampling Variability Tables for estimates of the number proportion or percentage of the surveyed population possessing a certain characteristic and for ratios and differences between such estimates Rule 1 Estimates of Numbers of Persons Possessing a Characteristic Aggregates The coefficient of variation depends only on the size of the estimate itself On the Approximate Sampling Variability Table for the appropriate geographic area locate the estimated number in the left most column of the table headed Numerator of Percentage and follow the asterisks if any across to the first figure encountered This figure is the approximate coefficient of variation Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide Rule2 Estimates of Proportions or Percentages of Persons Possessing a Characteristic The coefficient of variation of an estimated proportion or percentage depends on both th
25. and no records were lost or dropped Duplicate records are sometimes created due to transmission problems When this happens one of two identical records is dropped or if the duplicates are not absolutely identical the record with the most information is kept In the EICS duplicates were rarely found Editing Editing consists of modifying the data at the individual variable level The main type of editing carried out for the EICS data is called flow edits refer to Section 7 2 The reports produced by the flow edit system were thoroughly examined to detect potential errors introduced in processing This examination focussed on items with high incidence of Not stated answers and items where a valid answer was changed to a Valid skip or Not stated Very few situations could not be explained The verification process however revealed a number of response errors refer to Section 8 2 4 Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide Coding Industry and occupation were coded by a specially trained group of people which helped reduce the risk of coding errors Items unique to this survey are likely more subject to coding errors or inconsistent coding from year to year No specific measure of coding errors is available Derived Variables A large number of derived variables were created from the EICS collected data All derived variables were specified in decision tables For each variable the pr
26. ant during the survey reference week see definition of Mothers above Original sample Refers to the population targeted by the EICS before it was expanded to include all mothers of an infant The original survey types consisted of e Type 1 same as current e Type2 including part time mothers e Type 3 excluding mothers who have not worked in two years and Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide e Type 4 including mothers with a recent break in employment It is important to note that only the definition of Type 1 the unemployed has not changed since 1997 Reference week The sample used for this survey is selected from persons who have completed their participation in the LFS Although interviews are done three to seven weeks after the LFS interviews the reference week for the survey is the same as for the LFS Reference month The reference month refers to the month which contains the reference week This is the reference period for questions related to income Reference year For mothers the reference year is the 12 months prior to the birth or adoption of their child For the regular EICS population the reference year is the 12 month period ending with the reference month Working during the reference week Working during reference week refers to any work of an hour or longer duration performed for pay or profit Full time part time employment Full time emplo
27. askatchewan 2 55 193 23 969 Alberta 2 84 172 49 840 British Columbia 3 67 105 32 181 Western Provinces 3 55 470 105 990 Canada 4 03 1 248 364 843 Special Surveys Division 43 Employment Insurance Coverage Survey 2006 User Guide Province and Region for 5 MOTHER 0 and TYPE 3 Design Effect Sample Size Population Atlantic Provinces 4 37 587 182 588 Quebec 5 43 485 443 513 Ontario 6 70 826 804 828 Manitoba and Saskatchewan 4 43 378 122 240 Alberta 3 64 299 207 382 British Columbia 3 76 293 259 171 Western Provinces 4 99 970 588 793 Canada 6 90 2 868 2 019 723 Province and Region for MOTHER 0 and TYPE 2 or 4 Design Effect Sample Size Population Atlantic Provinces 6 97 622 222 557 Quebec 5 15 799 847 223 Ontario 6 13 1 343 1 431 907 Manitoba and Saskatchewan 3 75 743 277 588 Alberta 3 87 466 422 397 British Columbia 1 60 559 542 366 Western Provinces 2 62 1 768 1 242 351 Canada 5 07 4 532 3 744 037 All coefficients of variation in the Approximate Sampling Variability Tables are approximate and therefore unofficial Estimates of actual variance for specific variables may be obtained from Statistics Canada on a cost recovery basis Since the approximate CV is conservative the use of actual variance estimates may cause the estimate to be switched from one quality level to another For instance a margin
28. ated growth in relatively small areas whereas sample stabilization accommodates the slow sample growth over time that is the result of a fixed sampling rate along with a general increase in the size of the population Sample stabilization is the random dropping of dwellings from the sample in order to maintain the sample size at its desired level The basic weight is adjusted by the ratio of the sample size based on the fixed sampling rate to the desired sample size This adjustment factor is known as the stabilization weight The adjustment is done within stabilization areas defined as dwellings belonging to the same employment insurance economic region and the same rotation group Non response For certain types of non response i e household temporarily absent refusal data from a previous month s interview with the household if any is brought forward and used as the current month s data for the household In other cases non response is compensated for by proportionally increasing the weights of responding households The weight of each responding record is increased by the ratio of the number of households that should have been interviewed divided by the number that were actually interviewed This adjustment is done separately for non response areas which are defined by employment insurance economic region type of area and rotation group It is based on the assumption that the households that have been interviewed represent the characterist
29. bility of receiving benefits Such groups include e the long term jobless labour market entrants and students people becoming unemployed after uninsured employment people who have left jobs voluntarily and individuals who are eligible given their employment history but do not claim or otherwise receive benefits Employment Insurance coverage of the unemployed The survey data were used to classify individuals as either potentially eligible by El or not potentially eligible based on information provided by respondents about their claiming and receiving of benefits their perceived reasons for not receiving benefits or for not claiming and their recent labour market history The term potentially eligible for Employment Insurance is used here to describe unemployed people who during the reference week received El benefits or were in a position to receive them because of their recent insurable employment and subsequent job loss The term not potentially eligible describes the situation of those who did not receive benefits and could not have received them even if they had claimed as determined from the reported information The EICS provides an insight into the composition of the unemployed particularly those not receiving Employment Insurance benefits during the period of a reference week It provides a more meaningful picture of who does or does not have access to El benefits than do beneficiary unemployed B U ratio indicators
30. bles The codebook available for the public use microdata file refer to Chapter 13 0 includes a note that identifies all questionnaire items used to create each derived variable Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 7 5 1 Grouping of Continuous Data Items Most data items collected as continuous variables are only included on the public use microdata file as grouped variables Examples of such items are the age of the respondent AGECAT job tenure TENURE G number of weeks worked in the reference year WEEKSCAT and notice before job loss NOTICE W In other situations categorical response items were regrouped to create meaningful categories or to reduce the risk of identifying individuals with unique sets of answers This is the case for highest level of educational attainment EDUC industry and occupation NAICS6 and OCC6 job search methods JOBSRCH help needed in finding a job HELPFIND child care arrangements CHLDCARE household income range M HHINC for mothers only type of economic family EFAMILY and a few others 7 5 2 Combining Identical Questions Before 2004 the EICS questionnaire refer to Chapter 12 0 contained a number of questions which were duplicated two or thee times with slightly different wording This practice was frequent for questions related to claiming and receiving El benefits Different wording was used for mothers to reflect a different time
31. ccurring systematically will contribute to biases in the survey estimates Considerable time and effort were taken to reduce non sampling errors in the survey Quality assurance measures were implemented at each step of the data collection and processing cycle to monitor the quality of the data These measures include the use of highly skilled interviewers extensive training of interviewers with respect to the survey procedures and questionnaire observation of interviewers to detect problems of questionnaire design or misunderstanding of instructions procedures to ensure that data capture errors were minimized and coding and edit quality checks to verify the processing logic Special Surveys Division 31 32 Employment Insurance Coverage Survey 2006 User Guide 8 2 1 The Frame Because the EICS was a supplement to the Labour Force Survey LFS the frame used was the LFS sample Any non response to the LFS had an impact on the EICS frame The quality of the sampling variables in the frame was very high The EICS sample consisted of one rotation group from the LFS for the regular EICS population and of two rotation groups for mothers Note that the LFS frame excludes about 296 of all households in the 10 provinces of Canada Therefore the EICS frame also excludes the same proportion of households in the same geographical area It is unlikely that this exclusion introduces any significant bias into the survey data The EICS frame also e
32. cords that yield good aggregate estimates We can distinguish between three types of non response Complete non response is when the respondent does not provide the minimum set of answers These records are dropped and accounted for in the weighting process see Chapter 11 0 Item non response is when the respondent does not provide an answer to one question but goes on to the next question These are usually handled using the not stated code or are imputed Finally partial non response is when the respondent provides the minimum set of answers but does not finish the interview These records can be handled like either complete non response or multiple item non response Imputation was used to eliminate or reduce missing information caused by application problems in 2000 and 2001 This procedure was not repeated in subsequent years Users will find item specific information in the notes included in the survey master file codebooks There was no imputation done for the 2006 Employment Insurance Coverage Survey 7 5 Creation of Derived Variables A large number of data items on the microdata file have been derived by combining items on the questionnaire in order to facilitate data analysis All items on the public use microdata file were given a short name that abbreviates the variable description in English There are several types of derived variables on the data file This section provides general information about each type of derived varia
33. d incorporated unincorporated with without employees 3 Private unpaid family worker Type of work arrangement WRKTYP derived variable Coverage Respondents who ever worked 01 Permanent full time worker FTPT 1 and PERMTEMP 1 02 Permanent part time worker FTPT 2 and PERMTEMP 1 03 Permanent work hours unknown FTPT 9 and PERMTEMP 1 04 Not permanent seasonal worker PERMTEMP 2 05 Not permanent other PERMTEMP 3 4 or 5 06 Self employed COW 2 Other derived variables are created using more complex rules This is the case of COV a derived variable created to establish coverage of the El program 7 5 5 Taxonomy of Employment Insurance Coverage the COV Variable The EICS provides information on the situation of non working individuals relative to EI benefits It is a survey and not an administrative data source The El administrative data represents the actual decisions of Employment Insurance agents about benefit claims received by Human Resources and Social Development Canada HRSDC On the other hand in the EICS estimates of the degree of coverage of the Canadian population by the El program are made on the basis of behaviours events and perceptions reported by respondents in a household telephone survey The following is a description of the logic of the taxonomy used by HRSDC in reporting EI coverage of the unemployed The categories of coverage were determined in a hierarchical order described belo
34. d that these estimates flagged with the letter F do not meet Statistics Canada s quality standards Conclusions based on these data will be unreliable and most likely invalid Special Surveys Division 39 Employment Insurance Coverage Survey 2006 User Guide 9 6 Release Cut off s for the Employment Insurance Coverage Survey The following table provides an indication of the precision of population estimates as it shows the release cut offs associated with each of the three quality levels presented in the previous section These cut offs are derived from the coefficient of variation CV tables discussed in Chapter 10 0 For example the table shows that the quality of a weighted estimate of 40 000 people with Type 1 possessing a given characteristic in the Atlantic Provinces is marginal Note that these cut offs apply to estimates of total number of persons possessing a characteristic To estimate ratios users should not use the numerator value nor the denominator in order to find the corresponding quality level Rule 4 in Section 10 1 and Example 4 in Section 10 1 1 explain the correct procedure to be used for ratios Province and Region for Acceptable CV Marginal CV Unacceptable CV TYPE 1 0 0926 to 16 596 16 6926 to 33 396 gt 33 3 Atlantic Provinces 56 900 amp over 22 600 to lt 56 900 under 22 600 Quebec 102 700 8 over 33 400 to 102
35. das epe besote raden 18 5 54 OtherExCclusioris i o e atari Gis 19 5 6 Sample Size by Province for the Employment Insurance Coverage Survey 19 6 0 Data Collection EE 21 6 1 Interviewing for the Labour Force Survey sse 21 6 2 Supervision and Quality Control nanne ereen enne eeren nere eneeennneneneereneenanenenneenenenn 21 6 3 Non response to the Labour Force Survey unnnnsseesnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn nn 21 6 4 Data Collection Modifications for the Employment Insurance Coverage Survey 22 6 5 Non response to the Employment Insurance Coverage Survey nnn ennen eneen 22 7 0 Data Processing ireann ariia naran r NERENN E ENNEN E ASAE NAASE AAAA EAA 23 7 1 Data Capture nerna ce eet en HIE velat PD b Tod E nico 23 7 2 Verification and Editing iecit iei rea a ennn dend 23 7 3 Coding of Open ended Questions uurs40unnnennnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnannnnnnen 24 7 4 IMPUtAON cia ee leere leiser les 24 7 5 Creation of Derived Variables sssssssssssseseeee nennen nennen snnt nennen 24 7 5 1 Grouping of Continuous Data Items 25 7 5 2 Combining Identical Questions sss 25 7 5 8 Combining Data From the Labour Force Survey and the Employment Insurance Coverage Survey sssssssssssesee esent 25 7 5 4 Combining Two or More Different Questions 25 7 5 5 Taxonomy of Employment Insurance Co
36. des people who received or expect to receive El benefits in their current unemployment spell and individuals who have worked in a paid Special Surveys Division 13 Employment Insurance Coverage Survey 2006 User Guide job in the year prior to losing or leaving their last job and likely accumulated enough hours to qualify for El benefits Not potentially eligible for El This group includes unemployed persons without insurable employment in the last 12 months and also persons who quit their job without cause or in order to return to school Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 5 0 Survey Methodology The Employment Insurance Coverage Survey EICS has been administered since 1997 to a sub sample of the dwellings in the Labour Force Survey LFS sample and therefore its sample design is closely tied to that of the LFS The LFS design is briefly described in the Sections 5 1 to 5 4 Sections 5 5 and 5 6 describe how the EICS departed from the basic LFS design 5 1 Population Coverage The LFS is a monthly household survey of a sample of individuals who are representative of the civilian non institutionalized population 15 years of age or older in Canada s 10 provinces Specifically excluded from the survey s coverage are residents of the Yukon Northwest Territories and Nunavut persons living on Indian Reserves full time members of the Canadian Armed Forces and inmates of institutions Thes
37. e information is available on accepted claims but not for disallowed claims or for non claimants The administrative data also lacks demographic and household information which is necessary for social analysis The survey results fill several of these data gaps and allow users to draw a comprehensive profile of the unemployed and other persons who may have been entitled to El benefits due to a recent break in employment or a situation of underemployment The scope of the survey was broadened in 2000 to cover the access to maternity and parental benefits These changes were implemented one year before the expansion of the parental benefits program in January 2001 Special Surveys Division 7 Employment Insurance Coverage Survey 2006 User Guide 3 0 Objectives The primary objective of the Employment Insurance Coverage Survey EICS is to track the performance of the Employment Insurance EI program by finding out how many people are covered by EI what proportion of people receive benefits and which groups of people who may need El do not get access to Employment Insurance The data are used to measure the coverage of the Canadian population by Employment Insurance and the role El benefits play in contributing to personal and household income during periods of unemployment or underemployment The unemployed as well as working individuals e g beneficiaries with earnings and those categorized as not in the labour force by the Labour Fo
38. e reference week or the proportion of the unemployed eligible for El benefits are examples of such estimates An estimate of the number of persons possessing a certain characteristic may also be referred to as an estimate of an aggregate Examples of Categorical Questions Q Were Employment Insurance premiums deducted from your wages or salary at that job with employer name Yes No R Q What type of benefits did you receive that week R Training Regular Maternity only if female Parental Sickness Fishing Other 9 3 2 Quantitative Estimates Quantitative estimates are estimates of totals or of means medians and other measures of central tendency of quantities based upon some or all of the members of the surveyed population They also specifically involve estimates of the form X Y where X isan estimate of surveyed population quantity total and Y isan estimate of the number of persons in the surveyed population contributing to that total quantity An example of a quantitative estimate is the average number of months of leave taken from work after the birth or adoption of a child The numerator is an estimate of the total number of months of leave taken by all mothers for whom the information is available returned to work already or know plans and its denominator is the number of mothers taking leave of a known duration Examples of Quantitative Questions Q How long was this break from working in terms of
39. e groups together represent an exclusion of approximately 296 of the population aged 15 or over 5 2 Sample Design The LFS has undergone an extensive redesign culminating in the introduction of the new design at the end of 1994 The LFS sample is based upon a stratified multi stage design employing probability sampling at all stages of the design The design principles are the same for each province A diagram summarizing the design stages can be found in the document LFS_AppendixA pdf 5 2 1 Primary Stratification Provinces are divided into economic regions ER and employment insurance economic regions EIER ERs are geographic areas of more or less homogeneous economic structure formed on the basis of federal provincial agreements They are relatively stable overtime EIERs are also geographic areas and are roughly the same size and number as ERs but they do not share the same definitions Labour force estimates are produced for the EIERs for the use of Human Resources and Social Development Canada The intersections of the two types of regions form the first level of stratification for the LFS These ER EIER intersections are treated as primary strata and further stratification is carried out within them see Section 5 2 3 Note that a third set of regions census metropolitan areas CMA is also respected by stratification in the current LFS design since each CMA is also an EIER 5 2 2 Types of Areas The primary strata ER EIER
40. e instructed to make all reasonable attempts to obtain LFS interviews with members of eligible households For individuals who at first refuse to participate in the LFS a letter is sent from the Regional Office to the dwelling address stressing the importance of the survey and the household s cooperation This is followed by a second call or visit from the interviewer For cases in which the timing of the interviewer s call or visit is inconvenient an appointment is arranged to call back at a more convenient time For cases in which there is no one home numerous call backs are made Under no circumstances are sampled dwellings replaced by other dwellings for reasons of non response Special Surveys Division 21 22 Employment Insurance Coverage Survey 2006 User Guide Each month after all attempts to obtain interviews have been made a small number of non responding households remain For households non responding to the LFS and for which LFS information was obtained in the previous month this information is brought forward and used as the current month s LFS information No supplementary survey information is collected for these households 6 4 Data Collection Modifications for the Employment Insurance Coverage Survey Household members selected for the Employment Insurance Coverage Survey EICS are contacted three to seven weeks after their last LFS interview All interviews are conducted over the telephone and proxy response is n
41. e potentially eligible to receive El benefits were eligible to receive El benefits 10 3 How to Use the Coefficient of Variation Tables to Do a T test Standard errors may also be used to perform hypothesis testing a procedure for distinguishing between population parameters using sample estimates The sample estimates can be numbers averages percentages ratios etc Tests may be performed at various levels of significance where a level of significance is the probability of concluding that the characteristics are different when in fact they are identical Let X and be sample estimates for two characteristics of interest Let the standard error on the difference X X be 05 X If 1 222 is between 2 and 2 then no conclusion about the difference between the O d characteristics is justified at the 5 level of significance If however this ratio is smaller than 2 or larger than 2 the observed difference is significant at the 0 05 level That is to say that the difference between the estimates is significant Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 10 3 1 Example of Using the Coefficient of Variation Tables to Do a T test Let us suppose that the user wishes to test at 5 level of significance the hypothesis that there is no difference between the proportion of the regular employed population in Quebec who contributed to El and the proportion of the regular employed
42. e remaining cases were all included in the sample Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide Type 3 a full time students who left their last job because of school and full time students who did not leave their job more than one year ago were sub sampled at the rate of 50 b the remaining persons were all included Type 4 The sampling varied depending on the size of the stratum For 2006 it ranged from 15 to 100 Type 5 All persons were included 5 5 4 Other Exclusions At the second stage of sub sampling when three or more persons targeted by the EICS lived in the same household only two persons were selected into the survey unless they were all unemployed In this case a maximum of three persons were kept in the sample This was done to reduce the response burden within the household Some persons did not respond to the LFS interview they had imputed data or gave no permission to LFS personnel to conduct telephone interviews with them These were also excluded from the EICS 5 6 Sample Size by Province for the Employment Insurance Coverage Survey The following table shows the number of persons in the LFS sampled rotations that were selected in the EICS sample Province Sample Size Newfoundland and Labrador 697 Prince Edward Island 406 Nova Scotia 788 New Brunswick 738 Quebec 2 707 Ontario 4 206 Manitoba 1 007 Saskatchewan 1 039
43. e sample design itself determines weights which must be used to produce unbiased estimates of population Each record must be weighted by the inverse of the probability of selecting the person to whom the record refers In the example of a 296 simple random sample this probability would be 0 02 for each person and the records must be weighted by 1 0 02 50 Due to the complex LFS design dwellings in different regions will have different basic weights Because all eligible individuals in a dwelling are interviewed directly or by proxy this probability is essentially the same as the probability with which the dwelling is selected Cluster Sub weight The cluster delineation is such that the number of dwellings in the sample increases very slightly with moderate growth in the housing stock Substantial growth can be tolerated in an isolated cluster before the additional sample represents a field collection problem However if growth takes place in more than one cluster in an interviewer assignment the cumulative effect of all increases may create a workload problem In clusters where substantial growth has taken place sub sampling is used as a means of keeping interviewer assignments manageable The cluster sub weight represents the inverse of this sub sampling ratio in clusters where sub sampling has occurred Stabilization Weight Sample stabilization is also used to address problems with sample size growth Cluster sub sampling addressed isol
44. e size of the proportion or percentage and the size of the total upon which the proportion or percentage is based Estimated proportions or percentages are relatively more reliable than the corresponding estimates of the numerator of the proportion or percentage when the proportion or percentage is based upon a sub group of the population For example the proportion of unemployed receiving regular Employment Insurance EI benefits during the reference week is more reliable than the estimated number of unemployed receiving regular El benefits during the reference week Note that in the tables the coefficients of variation decline in value reading from left to right When the proportion or percentage is based upon the total population of the geographic area covered by the table the CV of the proportion or percentage is the same as the CV of the numerator of the proportion or percentage In this case Rule 1 can be used When the proportion or percentage is based upon a subset of the total population e g those in a particular sex or age group reference should be made to the proportion or percentage across the top of the table and to the numerator of the proportion or percentage down the left side of the table The intersection of the appropriate row and column gives the coefficient of variation Rule3 Estimates of Differences Between Aggregates or Percentages The standard error of a difference between two estimates is approximately equal to th
45. e square root of the sum of squares of each standard error considered separately That is the standard error of a difference X xx is Va 2307 o d X a X 20 where x is estimate 1 Ke is estimate 2 and and a are the coefficients of variation of X and X respectively The coefficient of variation of d is given by o d This formula is accurate for the difference between separate and uncorrelated characteristics but is only approximate otherwise Rule 4 Estimates of Ratios In the case where the numerator is a subset of the denominator the ratio should be converted to a percentage and Rule 2 applied This would apply for example to the case where the denominator is the number of unemployed potentially eligible for El and the numerator is the number of unemployed eligible for El In the case where the numerator is not a subset of the denominator as for example the ratio of the number of unemployed in receipt of regular El benefits as compared to the number of unemployed in receipt of any other type of benefits the standard error of the ratio of the estimates is approximately equal to the square root of the sum of squares of each coefficient of variation considered separately multiplied by R That is the standard error of a ratio R X c is D 2 2 o Rja a Special Surveys Division 45 Employment Insurance Coverage Survey 2006 User Guide where and a are the coefficients of variatio
46. ed into clusters and then a sample of clusters is selected within the stratum Dwellings are then sampled from selected clusters Different methods are used to define the clusters depending on the type of stratum Within each urban stratum in the urban area frame a number of geographically contiguous groups of dwellings or clusters are formed based upon 1991 Census counts These clusters are generally a set of one or more city blocks or block faces The selection of a sample of clusters always six or a multiple of six clusters from each of these secondary strata represents the first stage of sampling in most urban areas In some other urban areas census enumeration areas EA are used as clusters In the low density urban strata a three stage design is followed Under this design two towns within a stratum are sampled and then 6 or 24 clusters within each town are sampled For urban apartment strata instead of defining clusters the apartment building is the primary sampling unit Apartment buildings are sampled from the list frame with probability proportional to the number of units in each building Within each of the secondary strata in rural areas where necessary further stratification is carried out in order to reflect the differences among a number of socio economic characteristics within each stratum Within each rural stratum six EAs or two or three groups of EAs are sampled as clusters Special Surveys Division Employment
47. er data users The monthly LFS sample consists of approximately 60 000 dwellings After excluding dwellings found to be vacant dwellings demolished or converted to non residential uses dwellings containing only ineligible persons dwellings under construction and seasonal dwellings about 54 000 dwellings remain which are occupied by one or more eligible persons From these dwellings LFS information is obtained for approximately 102 000 civilians aged 15 or over 5 4 Sample Rotation The LFS follows a rotating panel sample design in which households remain in the sample for six consecutive months The total sample consists of six representative sub samples or panels and each month a panel is replaced after completing its six month stay in the survey Outgoing households are replaced by households in the same or a similar area This results in a five sixths month to month sample overlap which makes the design efficient for estimating month to month changes The rotation after six months prevents undue respondent burden for households that are selected for the survey Because of the rotation group feature it is possible to readily conduct supplementary surveys using the LFS design but employing less than the full size sample 5 5 Modifications to the Labour Force Survey Design for the Employment Insurance Coverage Survey The EICS is collected in four cycles each year For each cycle the EICS uses the rotation group that has just completed
48. erage then goes on to identify respondents who did not contribute to El and therefore are not potentially eligible for El COV 12 Respondent has never worked COV 11 Respondent last worked more than 12 months ago COV 10 Respondent was not a paid employee in their last job or stated that they did not contribute to El in their last job using WRKTYP and RNBENRW The classification continues with the remaining respondents who contributed to El but are not potentially eligible because of their reason for leaving their last job COV 9 Respondent reported not claiming or receiving benefits because they went to school or gave their reason for leaving their last job as going to school using RNBENRW or RSWORK COV 8 Respondent reported not claiming or receiving benefits because they quit their last job voluntarily and other respondents who indicated that they quit their last job For the remaining respondents about one in seven unemployed individuals the main task was to determine El eligibility based on hours worked in the year preceding the interruption of work The last three categories in the taxonomy of COV rest largely but not exclusively on a survey based estimate of insurable hours worked in the previous year This estimate takes into consideration the number of weeks worked in that year the number of weekly hours worked on average when working full time and hours worked on average when working part ti
49. es These variables are the province type of respondent sex and a grouping of employment insurance regions 4 A final adjustment is done using two external non overlapping independent sources Human Resources and Social Development Canada provides estimated counts for regular beneficiaries with and without earnings The other source is LFS data which provides estimated counts for unemployment not seasonally adjusted The adjustment is done within a calibration process which ensures that the estimates produced with the EICS data match the counts from the external sources The final calibrated weight is equal to the weight before the calibration multiplied by the factor necessary to calibrate to the applicable independent source The extended part of the EICS survey population comprised of the mothers of infants less than one year old is excluded from this calibration Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide The resulting weight WTPM is the final weight which appears on the EICS microdata file Special Surveys Division 57 Employment Insurance Coverage Survey 2006 User Guide 12 0 Questionnaires 12 1 The Labour Force Survey Questionnaire The Labour Force Survey questionnaire LFS QuestE pdf is used to collect information on the current and most recent labour market activity of all household members 15 years of age or older It includes questions on hours of work job tenure type
50. fficient of variation of d is given by o d 0 025 0 095 0 263 4 Sothe approximate coefficient of variation of the difference between the estimates is 26 396 The difference between the estimates is considered marginal and Statistics Canada recommends that this estimate be flagged with the letter E or some similar identifier and be accompanied by a warning to caution subsequent users about the high levels of error associated with the estimate Example 4 Estimates of Ratios Suppose that the user estimates that 543 846 of the regular employed population in Quebec contributed to El while 775 530 of the regular population in Ontario contributed to El The user is interested in comparing the estimate of Quebec versus that of Ontario in the form of a ratio How does the user determine the coefficient of variation of this estimate 1 First of all this estimate is a ratio estimate where the numerator of the estimate X is the number of employed individuals in Quebec who contributed to El The denominator of the estimate X is the number of employed individuals in Ontario who contributed to El 2 Refer to the coefficient of variation tables for MOTHER 0 and TYPE 2 or 4 QUEBEC and MOTHER 0 and TYPE 2 or 4 ONTARIO 3 The numerator of this ratio estimate is 543 846 The figure closest to it is 500 000 The coefficient of variation for this estimate is found by referring to the first non asterisk entry on that row name
51. he appropriate table the coefficient of variation of the estimate X and then using the following formula to convert to a confidence interval CI CI 1o X 1 a where is the determined coefficient of variation of X and t 1 if a 6896 confidence interval is desired t 1 6 if a 9096 confidence interval is desired t 2 if a 95 confidence interval is desired t 2 6 if a 9996 confidence interval is desired Note Release guidelines which apply to the estimate also apply to the confidence interval For example if the estimate is not releasable then the confidence interval is not releasable either Special Surveys Division 51 Employment Insurance Coverage Survey 2006 User Guide 10 2 1 Example of Using the Coefficient of Variation Tables to Obtain Confidence Limits A 95 confidence interval for the estimated proportion of unemployed individuals who were potentially eligible to receive El benefits were eligible to receive El benefits from Example 2 Section 10 1 1 would be calculated as follows X 81 8 or expressed as a proportion 0 818 t 2 a 1 4 0 014 expressed as a proportion is the coefficient of variation of this estimate as determined from the tables CI 0 818 2 0 818 0 014 0 818 2 0 818 0 014 CI 0 818 0 023 0 818 0 023 CI 0 795 0 841 With 95 confidence it can be said that between 79 5 and 84 196 of unemployed individuals who wer
52. he level of confidence that the true value for the population lies within a specified range of values For example a 95 confidence interval can be described as follows If sampling of the population is repeated indefinitely each sample leading to a new confidence interval for an estimate then in 95 of the samples the interval will cover the true population value Using the standard error of an estimate confidence intervals for estimates may be obtained under the assumption that under repeated sampling of the population the various estimates obtained for a population characteristic are normally distributed about the true population value Under this assumption the chances are about 68 out of 100 that the difference between a sample estimate and the true population value would be less than one standard error about 95 out of 100 that the difference would be less than two standard errors and about 99 out of 100 that the difference would be less than three standard errors These different degrees of confidence are referred to as the confidence levels Confidence intervals for an estimate X are generally expressed as two numbers one below the estimate and one above the estimate as X k X k where k is determined depending upon the level of confidence desired and the sampling error of the estimate Confidence intervals for an estimate can be calculated directly from the Approximate Sampling Variability Tables by first determining from t
53. ics of those that should have been interviewed within a non response area Special Surveys Division 55 56 Employment Insurance Coverage Survey 2006 User Guide Labour Force Survey Sub weight The product of the previously described weighting factors is called the LFS sub weight All members of the same sampled dwelling have the same sub weight Sub provincial and Province Age Sex Adjustments The sub weight can be used to derive a valid estimate of any characteristic for which information is collected by the LFS However these estimates will be based on a frame that contains some information that may be several years out of date and therefore not representative of the current population Through the use of more up to date auxiliary information about the target population the sample weights are adjusted to improve both the precision of the estimates and the sample s representation of the current population Independent estimates are available monthly for various age and sex groups by province These are population projections based on the most recent census data records of births and deaths and estimates of migration In the final step this auxiliary information is used to transform the sub weight into the final weight This is done using a calibration method This method ensures that the final weights it produces sum to the census projections for the auxiliary variables namely totals for various age sex groups economic regions census
54. ime employed and those not in the labour force during the reference week who have not worked for two years were the principal exclusions 5 5 2 Type 4 A Special Case Respondents sampled with Type 4 are not all targeted by the survey Only those who have experienced an interruption in work in the two months prior to the survey reference week need to be interviewed This information was not available from the LFS sample frame Therefore all full time workers with short job tenure at their current job were selected The question on work interruption is asked in the EICS and respondents who worked continually over the two month period prior to the reference week are not asked further questions They are out of scope for the survey and their records are dropped in processing refer to Section 7 2 In a year roughly 40 of those selected with Type 4 are dropped for this reason 5 5 3 Sub sampling At the initial stage sub sampling was done to arrive at the target sample of 3 600 and to balance the representation of groups according to the relevance of the Employment Insurance program to them The sub sampling criteria are summarized by focus type as follows Type 1 All persons were included Type 2 a full time students were sub sampled at the rate of 7096 b persons working 20 to 24 hours during the reference week C persons working 25 to 29 hours during the reference week were sub sampled at the rate of approximately 50 d th
55. intersections are further disaggregated into three types of areas rural urban and remote areas Urban and rural areas are loosely based on the Census definitions of urban and rural with some exceptions to allow for the formation of strata in some areas Urban areas include the largest CMAs down to the smallest villages categorized by the 1991 Census as urban 1 000 people or more while rural areas are made up of areas not designated as urban or remote All urban areas are further subdivided into two types those using an apartment list frame and an area frame as well as those using only an area frame A detailed description of the LFS design is available in the Statistics Canada publication entitled Methodology of the Canadian Labour Force Survey Catalogue no 71 526 XPB Special Surveys Division 15 Employment Insurance Coverage Survey 2006 User Guide Approximately 196 of the LFS population is found in remote areas of provinces which are less accessible to LFS interviewers than other areas For administrative purposes this portion of the population is sampled separately through the remote area frame Some populations not congregated in places of 25 or more people are excluded from the sampling frame 5 2 3 Secondary Stratification In urban areas with sufficiently large numbers of apartment buildings the strata are subdivided into apartment frames and area frames The apartment list frame is a register maintained for the
56. ion There was no imputation of data to compensate for total or item non response in the EICS 8 2 5 Measurement of Sampling Error Since it is an unavoidable fact that estimates from a sample survey are subject to sampling error sound statistical practice calls for researchers to provide users with some indication of the magnitude of this sampling error This section of the documentation outlines the measures of sampling error which Statistics Canada commonly used and which it urges users producing estimates from this microdata file to use also The basis for measuring the potential size of sampling errors is the standard error of the estimates derived from survey results However because of the large variety of estimates that can be produced from a survey the standard error of an estimate is usually expressed relative to the estimate to which it pertains This resulting measure known as the coefficient of variation CV of an estimate is obtained by dividing the standard error of the estimate by the estimate itself and is expressed as a percentage of the estimate Special Surveys Division 33 34 Employment Insurance Coverage Survey 2006 User Guide For example suppose that based upon the 2003 EICS results one estimates that 8196 of individuals are eligible for Employment Insurance among the potentially eligible and this estimate is found to have a standard error of 0 03 Then the coefficient of variation of the estimate is
57. ios for qualitative variables can be derived using the accompanying Approximate Sampling Variability Tables Special Surveys Division 37 Employment Insurance Coverage Survey 2006 User Guide 38 For other analysis techniques for example linear regression logistic regression and analysis of variance a method exists which can make the variances calculated by the standard packages more meaningful by incorporating the unequal probabilities of selection The method rescales the weights so that there is an average weight of 1 For example suppose that analysis of all male respondents is required The steps to rescale the weights are as follows 1 select all respondents from the file who reported SEX men 2 calculate the AVERAGE weight for these records by summing the original person weights from the microdata file for these records and then dividing by the number of respondents who reported SEX men 3 for each of these respondents calculate a RESCALED weight equal to the original person weight divided by the AVERAGE weight 4 perform the analysis for these respondents using the RESCALED weight However because the stratification and clustering of the sample s design are still not taken into account the variance estimates calculated in this way are likely to be under estimates The calculation of more precise variance estimates requires detailed knowledge of the design of the survey Such detail cannot be given in this
58. irst calculate the approximate coefficient of variation for the Manitoba and Saskatchewan ratio R and the Alberta ratio R as in Example 4 The approximate CV for the Manitoba and Saskatchewan ratio is 12 896 and 11 396 for Alberta Using Rule 3 the standard error of a difference d R R is where a and are the coefficients of variation of R and R respectively That is the standard error of the difference d 1 24 1 21 0 03 is c J 1 24X0 128 F 1 210 113 0 025192 0 018695 0 209 The coefficient of variation of d is given by o d 0 209 0 03 6 967 So the approximate coefficient of variation of the difference between the estimates is 696 7 The difference between the estimates is considered unacceptable and Statistics Canada recommends this estimate not be released However should the user choose to do so the estimate should be flagged with the letter F or some Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide similar identifier and be accompanied by a warning to caution subsequent users about the high levels of error associated with the estimate 10 2 How to Use the Coefficient of Variation Tables to Obtain Confidence Limits Although coefficients of variation are widely used a more intuitively meaningful measure of sampling error is the confidence interval of an estimate A confidence interval constitutes a statement on t
59. its six months in the LFS The EICS collection follows the LFS collection for the months of March June October and December This sample is augmented by a second rotation for each cycle for mothers of infants Special Surveys Division 17 Employment Insurance Coverage Survey 2006 User Guide The survey estimates are produced for the reference year by averaging over the four cycles covered by the survey 5 5 1 Target Population The target population for this survey is a subpopulation of the LFS and focuses on five groups or types of persons who are potential employment insurance recipients 1 persons who were unemployed during the reference week persons employed part time during the reference week persons not in the labour force during the reference week persons employed full time during the reference week who started their current job during the previous three months 5 mothers of infants less than one year old working during the reference week Ron Of most relevance are the unemployed and the jobless but part time workers can also receive benefits e g if they recently had an interruption in earnings and are entitled to retain Employment Insurance El benefits while working due to small employment earnings One rotation group from the LFS typically includes approximately 5 500 individuals falling in one of the five target groups out of a total sample of approximately 22 000 individuals aged 15 and over Full t
60. lated variables were grouped for the EICS public use microdata file 7 5 4 Combining Two or More Different Questions Variables such as union status UNIONCA type of work arrangement WRKTYP reason stopped working RSWORK claiming benefits CLAIM receipt of benefits BENEFIT reason for not receiving or claiming benefits RNBENRW receipt of additional payments ADDPAYM and looking for work outside the community LOOKOUT are derived using more than one questionnaire item In these cases the algorithm used to create the new variable is usually fairly intuitive For instance the variable on type of work arrangements is created by combining full time or Special Surveys Division 25 26 Employment Insurance Coverage Survey 2006 User Guide part time status permanent or temporary employment status and reason for temporary employment and class of worker as follows Full time or part time status FTPT Coverage Paid employees at last or current job 1 Full time 2 Part time Permanent or temporary job status PERMTEMP only available on Master file Coverage Paid employees at last or current job 1 Permanent Not permanent seasonal job Not permanent temporary term or contract job Not permanent casual job Not permanent work done through a temporary help agency Not permanent other ooR WP Class of worker at main job COW Coverage Respondents who ever worked 1 Public or private employee 2 Self employe
61. ly 2 5 6 4 The denominator of this ratio estimate is 775 530 The figure closest to it is 750 000 The coefficient of variation for this estimate is found by referring to the first non asterisk entry on that row namely 2 5 5 Sothe approximate coefficient of variation of the ratio estimate is given by Rule 4 which is Special Surveys Division 49 50 Employment Insurance Coverage Survey 2006 User Guide 2 2 a a 0 where a and a are the coefficients of variation of X and X respectively That is a 0 025 0 025 0 000625 0 000625 0 035 The obtained ratio of Quebec versus Ontario individuals in the regular employed population contributing to El is 543 846 775 530 which is 0 70 to be rounded according to the rounding guidelines in Section 9 1 The coefficient of variation of this estimate is 3 5 which makes the estimate releasable with no qualifications Example 5 Estimates of Differences of Ratios Suppose that the user estimates that the ratio of individuals aged 15 to 24 years in the regular employed population who contributed to El to individuals aged 25 to 44 years in the regular employed population who contributed to El is 1 24 for Manitoba and Saskatchewan while it is 1 21 for Alberta The user is interested in comparing the two ratios to see if there is a statistical difference between them How does the user determine the coefficient of variation of the difference 1 F
62. me Usual hours worked in the most recent job or average hours worked for all part timers and full timers are used in case of non response The entrance criterion is set at 700 hours for all the highest entrance criteria across the country Special Surveys Division 27 Employment Insurance Coverage Survey 2006 User Guide 28 COV 7 Respondent reported not claiming or receiving El benefits because of a lack of sufficient hours of insurable work or because they had no recent work using RNBENRW Respondents whose tenure at the last job was less than or equal to three months since no information is available on the insurability of the hours worked at previous jobs within the year could have been self employment or other uninsured employment using TENURE C Survey estimate of insurable hours is less than 700 hours COV 5 Survey estimate of insurable hours is 700 or greater but respondent did not claim El benefits COV 6 Survey estimate of insurable hours is 700 or greater and respondent claimed El benefits did not receive This concludes the definition of COV The derived variable ELIGIBLE summarises COV as follows 1 Potentially eligible eligible COV 1 to 6 2 Potentially eligible not eligible COV 7 3 Not potentially eligible COV 8 to 12 The main measure of El coverage published from this survey expresses the estimate of eligible ELIGIBLE 1 as a percentage of potentially eligible ELIGIBLE 1 or 2
63. months _ _ months you work per week R Q During the weeks that you worked full time how many hours on average did R LI J hours Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 9 3 3 Tabulation of Categorical Estimates Estimates of the number of people with a certain characteristic can be obtained from the microdata file by summing the final weights of all records possessing the characteristic s of interest Proportions and ratios of the form X Y are obtained by a summing the final weights of records having the characteristic of interest for the numerator X b summing the final weights of records having the characteristic of interest for the denominator Y then C dividing estimate a by estimate b X Y 9 3 4 Tabulation of Quantitative Estimates Estimates of quantities can be obtained from the microdata file by multiplying the value of the variable of interest by the final weight for each record then summing this quantity over all records of interest For example to obtain an estimate of total number of weeks of Employment Insurance EI received by mothers of an infant who have already returned to work multiply the value reported in derived variable BENWEEKS weeks received El by the final weight for the record then sum this value over all records with MOTHER 1 and WORKNOW 1 mother of an infant less than one year old who are currently working To obt
64. n of x and x respectively The coefficient of variation of R is given by c l R The formula will tend to overstate the error if X and x are positively correlated and understate the error if x and x are negatively correlated Rule 5 Estimates of Differences of Ratios In this case Rules 3 and 4 are combined The CVs for the two ratios are first determined using Rule 4 and then the CV of their difference is found using Rule 3 10 1 1 Examples of Using the Coefficient of Variation Tables for Categorical Estimates The following examples based on the EICS 2008 are included to assist users in applying the foregoing rules Please note that the data for these examples are different than the results obtained from the current survey and are only to be used as a guide Example 1 Estimates of Numbers of Persons Possessing a Characteristic Aggregates Suppose that a user estimates that 400 393 unemployed individuals received regular El benefits during the reference week How does the user determine the coefficient of variation of this estimate 1 Refer to the coefficient of variation table for TYPE 1 CANADA Employment Insurance Coverage Survey 2000 to 2003 Approximate Sampling Variability Tables TYPE 1 Canada NUMERATOR OF ESTIMATED PERCENTAGE PERCENTAGE 000 0 1 2 0 5 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 50 0 98 9 979 964 93 9 91 2 885 87 828 798 766 70 0 P 69 3 682 664 645 626 606 585 564 542 49 5 dui 56
65. not at work due to factors such as own illness or disability personal or family responsibilities vacation labour dispute or other reasons excluding persons on layoff between casual jobs and those with a job to start at a future date Unemployment Unemployed persons are those who during the reference week a were on temporary layoff during the reference week with the expectation of recall and were available for work or b were without work had actively looked for work in the past four weeks and were available for work or c had a new job to start within four weeks from the reference week and were available for work Not in the Labour Force Persons not in the labour force are those who during the reference week were unwilling or unable to offer or supply labour services under conditions existing in their labour markets that is they were neither employed nor unemployed Work includes any work for pay or profit that is paid work in the context of an employer employee relationship or self employment It also includes unpaid family work which is defined as unpaid work contributing directly to the operation of a farm business or professional practice owned and operated by a related member of the same household Such activities may include keeping books selling products waiting on tables and so on Tasks such as housework or maintenance of the home are not considered unpaid family work N Persons are regarded as a
66. o 00 and the preceding digit the hundreds digit is left unchanged If the last digits are between 50 and 99 they are changed to 00 and the preceding digit is incremented by 1 b Marginal sub totals and totals in statistical tables are to be derived from their corresponding unrounded components and then are to be rounded themselves to the nearest 100 units using normal rounding C Averages proportions rates and percentages are to be computed from unrounded components i e numerators and or denominators and then are to be rounded themselves to one decimal using normal rounding In normal rounding to a single digit if the final or only digit to be dropped is O to 4 the last digit to be retained is not changed If the first or only digit to be dropped is 5 to 9 the last digit to be retained is increased by 1 d Sums and differences of aggregates or ratios are to be derived from their corresponding unrounded components and then are to be rounded themselves to the nearest 100 units or the nearest one decimal using normal rounding e In instances where due to technical or other limitations a rounding technique other than normal rounding is used resulting in estimates to be published or otherwise released which differ from corresponding estimates published by Statistics Canada users are urged to note the reason for such differences in the publication or release document s f Under no circumstances are unrounded estimates to be p
67. ocess generates a summary table documenting the rules applied and rule counts The distribution for each derived variable was compared to that of the questionnaire items used in creating it The derived variables were also cross classified with other related variables to ensure internal consistency and limit the risk of errors in the derivation rules A comparison of the distribution over the 2005 and 2006 period was also conducted to ensure historical comparability of the information included on the public use microdata files 8 2 4 Non response A major source of non sampling errors in surveys is the effect of non response on the survey results The extent of non response varies from partial non response failure to answer just one or some questions to total non response Total non response occurred because the interviewer was either unable to contact the respondent or the respondent refused to participate in the survey Total non response was handled by adjusting the weight of individuals who responded to the survey to compensate for those who did not respond It was consistently more pronounced among the full time employed Type 4 or 5 over the years and also marginally for men but there is no marked difference across broad age groups In most cases partial item non response to the survey occurred when the respondent did not understand or misinterpreted a question refused to answer a question or could not recall the requested informat
68. of work reason for hours lost or absent job search undertaken availability for work and school attendance 12 2 The Employment Insurance Coverage Survey Questionnaire The Employment Insurance Coverage Survey EICS questionnaire was used in 2006 to collect the information for the supplementary survey The file EICS2006 QuestE pdf contains the English questionnaire Special Surveys Division 59 Employment Insurance Coverage Survey 2006 User Guide 13 0 Record Layout with Univariate Frequencies See EICS2006 Master CdBk pdf for the record layout with univariate counts Special Surveys Division 61
69. or weighted estimates based on sample sizes of 30 or more users should determine the coefficient of variation of the estimate and follow the guidelines below These quality level guidelines should be applied to rounded weighted estimates All estimates can be considered releasable However those of marginal or unacceptable quality level must be accompanied by a warning to caution subsequent users Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide Quality Level Guidelines Quality Level of Estimate Guidelines 1 Acceptable Estimates have a sample size of 30 or more and low coefficients of variation in the range of 0 0 to 16 5 No warning is required 2 Marginal Estimates have a sample size of 30 or more and high coefficients of variation in the range of 16 6 to 33 3 Estimates should be flagged with the letter E or some similar identifier They should be accompanied by a warning to caution subsequent users about the high levels of error associated with the estimates 3 Unacceptable Estimates have a sample size of less than 30 or very high coefficients of variation in excess of 33 3 Statistics Canada recommends not to release estimates of unacceptable quality However if the user chooses to do so then estimates should be flagged with the letter F or some similar identifier and the following warning should accompany the estimates Please be warne
70. ot allowed in the EICS There may be more than one person selected in each household but never more than three 6 5 Non response to the Employment Insurance Coverage Survey Similar to the LFS the interviewers are asked to make all reasonable efforts to obtain the EICS interview Refusals at first contact are followed up by a senior interviewer However contrary to the LFS no letters are sent to help obtain the respondent s cooperation Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide 7 0 Data Processing The main output of the Employment Insurance Coverage Survey EICS is a clean microdata file This chapter presents a brief summary of the processing steps involved in producing this file 7 1 Data Capture Responses to survey questions are captured directly by the interviewer at the time of the interview using a computerized questionnaire The computerized questionnaire reduces processing time and costs associated with data entry transcription errors and data transmission The response data are encrypted to ensure confidentiality and sent via modem to the appropriate Statistics Canada Regional Office From there they are transmitted over a secure line to Ottawa for further processing Some editing is done directly at the time of the interview Where the information entered is out of range too large or small of expected values or inconsistent with the previous entries the interviewer is prompted
71. rce Survey LFS are the objects of analysis under this topic The latter two groups also receive Employment Insurance benefits in significant numbers The factors cited most frequently to explain variations in El coverage are not qualifying for El exhausting benefits serving a waiting period after job separation or not claiming El The magnitude of these and other factors and their correlation to personal characteristics seasonal and business cycles and regions of Canada can be investigated using this survey to improve our understanding of the reasons why some unemployed do not receive El benefits Through the survey data analysts will also be able to observe the characteristics and situation of people not covered by El and of those who exhausted EI benefits the job search intensity of the unemployed expectation of recall to a job and alternate sources of income and funds Survey data pertaining to maternity and parental benefits answer questions on the proportion of mothers of an infant who received maternity and parental benefits the reason why they don t and about sharing parental benefits with their spouse The survey also allows looking at the timing and circumstances related to the return to work the income adequacy of households with young children and more The Employment Insurance Coverage Survey The survey was designed to produce a series of precise measures of the unemployed population in order to identify groups with low proba
72. rly earnings HRLYEARN which does include a small percentage of outliers and internal consistency working individuals reporting zero earnings 7 3 Coding of Open ended Questions A few data items on the questionnaire were recorded by interviewers in an open ended format In the EICS the coding process assigns standard codes to the industry and occupation descriptions provided by the respondents North American Industry Classification System NAICS and the Standard Occupational Classification SOC 91 and to the country of birth Also Other specify fields with a significant number of text answers were examined and coded to existing categories In some occasions new categories were created to facilitate the analyses of the textual information These were items relating to reasons for not claiming or receiving benefits industry occupation reason for interrupting work job search method used reason why spouse did not claim benefits or why both parents did 7 4 Imputation Imputation is the process that supplies valid values for those variables that have been identified for a change either because of invalid information or because of missing information The new values are supplied in such a way as to preserve the underlying structure of the data and to ensure that the resulting records will pass all required edits In other words the objective is not to reproduce the true microdata values but rather to establish internally consistent data re
73. roduced These CV tables allow the user to obtain an approximate coefficient of variation based on the size of the estimate calculated from the survey data The coefficients of variation are derived using the variance formula for simple random sampling and incorporating a factor which reflects the multi stage clustered nature of the sample design This factor known as the design effect was determined by first calculating design effects for a wide range of characteristics and then choosing from among these a conservative value to be used in the CV tables which would then apply to the entire set of characteristics The table below shows the conservative value of the design effects as well as sample sizes and population counts by province for survey type and mother status which were used to produce the Approximate Sampling Variability Tables for the Employment Insurance Coverage Survey EICS Province and Region for Design Effect Sample Size Population TYPE z 1 Atlantic Provinces 14 59 522 112 525 Quebec 5 78 438 314 800 Ontario 6 07 559 407 050 Manitoba and Saskatchewan 3 86 192 43 788 Alberta 3 64 106 56 863 British Columbia 3 61 160 104 399 Western Provinces 4 55 458 205 050 Canada 2 18 1 977 1 039 425 Province and Region for 3 MOTHER 1 Design Effect Sample Size Population Atlantic Provinces 1 99 167 21 136 Quebec 4 38 227 89 903 Ontario 3 74 384 147 814 Manitoba and S
74. the following week called the Survey Week and the labour force status determined is that of the reference week Full time Employment Full time employment consists of persons who usually work 30 hours or more per week at their main or only job Part Time Employment Part time employment consists of persons who usually work less then 30 hours per week at their main or only job 42 Employment Insurance Coverage Survey Concepts and Definitions Type The EICS sample represents five distinct subpopulations of interest called Type Type is defined as follows 1 persons who were unemployed during the reference week persons employed part time during the reference week persons not in the labour force during the reference week persons employed full time during the reference week who started their current job during the previous three months 5 mothers of infants less than one year old working during the reference week The type often determines which questions are asked in the survey AUN Mothers In this survey the term mother refers to mothers by birth or adoption of an infant aged less than one year old during the LFS reference week Many mothers were not part of the survey sample prior to 2000 In particular mothers working full time and mothers not in the labour force and who have not worked in the past two years or ever were not included in the survey prior to 2000 Regular population Not the mother of an inf
75. tia 658 554 84 New Brunswick 608 531 87 Quebec 2 270 1 946 86 Ontario 3 630 3 104 86 Manitoba 836 724 87 Saskatchewan 887 777 88 Alberta 1 267 1 042 82 British Columbia 1 301 1 115 86 Canada 12 407 10 601 85 Note The EICS response rate is the number of EICS responding individuals as a percentage of the number of EICS selected individuals in scope refer to Sections 5 5 2 and 7 2 8 2 Survey Errors The estimates derived from this survey are based on a sub sample of individuals from the Labour Force Survey Somewhat different estimates might have been obtained if a complete census had been taken using the same questionnaire interviewers supervisors processing methods etc as those actually used in the survey The difference between the estimates obtained from the sample and those resulting from a complete count taken under similar conditions is called the sampling error of the estimate Errors which are not related to sampling may occur at almost every phase of a survey operation Interviewers may misunderstand instructions respondents may make errors in answering questions the answers may be incorrectly entered on the questionnaire and errors may be introduced in the processing and tabulation of the data These are all examples of non sampling errors Over a large number of observations randomly occurring errors will have little effect on estimates derived from the survey However errors o
76. tical Analysis enne nnns 37 9 5 Coefficient of Variation Release Guidelines sse 38 9 6 Release Cut off s for the Employment Insurance Coverage Survey 40 Approximate Sampling Variability Tables eeceeeeeeeeeeeee nien nnne nnne nnne nnnm nnns 43 10 1 Howto Use the Coefficient of Variation Tables for Categorical Estimates 44 10 1 1 Examples of Using the Coefficient of Variation Tables for Categorical Estimates rue ep RE nce a Ox Pe og a ee 46 10 2 Howto Use the Coefficient of Variation Tables to Obtain Confidence Limits 51 10 2 1 Example of Using the Coefficient of Variation Tables to Obtain Confidence LINS ice ic ec nne pete eani Genes 52 10 3 Howto Use the Coefficient of Variation Tables to Do a T test nennen nennen eenn 52 10 3 1 Example of Using the Coefficient of Variation Tables to Do a T test 53 10 4 Coefficients of Variation for Quantitative Estimates sssssseeeee 53 10 5 Coefficient of Variation Tables ssssssssssseseseee eene enne entente nnne nnne nennen 53 Weighting ensem eiecti 55 11 1 Weighting Procedures for the Labour Force Survey ssssssseeee 55 11 2 Weighting Procedures for the Employment Insurance Coverage Survey 56 Questionnaires 20 59 12 1 The
77. tion collected in the first month and collects the labour force information for the current month In each dwelling information about all household members is usually obtained from one knowledgeable household member Such proxy reporting which accounts for approximately 65 of the information collected is used to avoid the high cost and extended time requirements that would be involved in repeat visits or calls necessary to obtain information directly from each respondent If during the course of the six months that a dwelling normally remains in the sample an entire household moves out and is replaced by a new household information is obtained about the new household for the remainder of the six month period At the conclusion of the LFS monthly interviews interviewers introduce the supplementary survey if any to be administered to some or all household members that month 6 2 Supervision and Quality Control All LFS interviewers are under the supervision of a staff of senior interviewers who are responsible for ensuring that interviewers are familiar with the concepts and procedures of the LFS and it s many supplementary surveys and also for periodically monitoring their interviewers and reviewing their completed documents The senior interviewers are in turn under the supervision of the LFS program managers located in each of the Statistics Canada regional offices 6 3 Non response to the Labour Force Survey Interviewers ar
78. ublished or otherwise released by users Unrounded estimates imply greater precision than actually exists 9 2 Sample Weighting Guidelines for Tabulation The sample design used for the Employment Insurance Coverage Survey EICS was not self weighting When producing simple estimates including the production of ordinary statistical tables users must apply the proper survey weights If proper weights are not used the estimates derived from the microdata files cannot be considered to be representative of the survey population and will not correspond to those produced by Statistics Canada Special Surveys Division 35 36 Employment Insurance Coverage Survey 2006 User Guide Users should also note that some software packages may not allow the generation of estimates that exactly match those available from Statistics Canada because of their treatment of the weight field 9 3 Definitions of Types of Estimates Categorical and Quantitative Before discussing how the EICS data can be tabulated and analyzed it is useful to describe the two main types of point estimates of population characteristics which can be generated from the microdata file for the EICS 9 39 1 Categorical Estimates Categorical estimates are estimates of the number or percentage of the surveyed population possessing certain characteristics or falling into some defined category The number of unemployed who received Employment Insurance EI benefits during th
79. vailable for work if they i reported that they could have worked in the reference week if a suitable job had been offered or if the reason they could not take a job was of a temporary nature such as because of own illness or disability personal or family responsibilities because they already have a job to start in the near future or because of vacation prior to 1997 those on vacation were not considered available ii were full time students seeking part time work who also met condition i above Full time students currently attending school and looking for full time work are not considered to be available for work during the reference week Special Surveys Division 11 Employment Insurance Coverage Survey 2006 User Guide Industry and Occupation The Labour Force Survey provides information about the occupation and industry attachment of employed and unemployed persons and of persons not in the labour force who have held a job in the past 12 months Since 1997 these statistics have been based on the North American Industry Classification System NAICS and the Standard Occupational Classification SOC 91 Prior to 1997 the 1980 Standard Industrial Classification and the 1980 Standard Occupational Classification were used Reference Week The entire calendar week from Sunday to Saturday covered by the Labour Force Survey each month It is usually the week containing the 15 day of the month The interviews are conducted during
80. verage the COV Variable 26 7 6 flee 28 7 7 Suppression of Confidential Information esssseseseseeeeen ene 28 Special Surveys Division 3 8 0 9 0 10 0 12 0 13 0 Employment Insurance Coverage Survey 2006 User Guide Data Quality tt 31 8 1 Response Hates c ac etn enten ee entente 31 8 2 SUWVey EMOS uti 31 8 2 1 The Frame cai A Heuer PE aea era re E RA engen 32 8 2 2 Data Collection nnen ennen nennen rr 32 8 23 Data Processing e eiecit ne einen 32 8 2 4 Nori feSpOnSB ideft eei iei eire cien etre eta deal ea ecdesia 33 8 2 5 Measurement of Sampling Error sse 33 Guidelines for Tabulation Analysis and Release unsunssasnnnnnnnnnnnnnnnannnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn 35 9 1 Rounding Guidelines tir ee 35 9 2 Sample Weighting Guidelines for Tabulation sseeeeenneenes 35 9 3 Definitions of Types of Estimates Categorical and Quantitative 36 9 3 1 Categorical Estimates ssesssssssssssseseese esee entree nennen nnne nen nennen 36 9 3 2 Quantitative Estimates sssssssssssssseseee eene nennen nnne nennen 36 9 3 8 Tabulation of Categorical Estimates ssssssssssee eee 37 9 3 4 Tabulation of Quantitative Estimates ssssssssseeeeennnens 37 9 4 Guidelines for Statis
81. w The first four categories are mutually exclusive and regroup all respondents who have received benefits since they last worked or expected to receive benefits for the reference week when interviewed Some respondents in these four groups have left their job returned to school were self employed in their last job or without work for more than one Special Surveys Division Employment Insurance Coverage Survey 2006 User Guide year Despite these circumstances the fact that they have received EI benefits in the past year clearly establishes their eligibility COV 1 Respondent received regular El benefits in the reference week using BENEFIT and BENTYP COV 2 Respondent received special El benefits in the reference week using BENEFIT and BENTYP COV 3 Respondent did not receive benefits during the reference week but expects to receive benefits in the non working period using BENEFIT and RNBENRW Persons are considered to be in a position of receiving benefits when they indicate that they claimed El benefits and say that they did not receive El benefits during the reference week but are still expecting benefit payments for that week or are serving a waiting period or benefits are being withheld due to severance or other payments or other reasons COV 4 Respondent did not receive benefits for the reference week but received some EI benefits since he she last worked in the last 12 months The taxonomy of the El cov
82. xcludes full non response to the LFS and item non response to variables used in the selection criteria The variables on the EICS frame were quite up to date since they were collected from the LFS at most three weeks before the beginning of the EICS collection 8 2 2 Data Collection Interviewer training consisted of reading the EICS Interviewer s Manual practicing with the EICS training cases on the computer and discussing any questions with senior interviewers before the start of the survey A description of the background and objectives of the survey was provided as well as a glossary of terms and a set of questions and answers Interviewers started collecting the EICS information two weeks after the end of the January April July and November LFS collection period Collection lasted five weeks for each EICS cycle 8 2 3 Data Processing Data processing of the EICS was done in a number of steps including verification coding editing estimation confidentiality etc At each step a picture of the output files is taken and a report showing changes to each variable from one step to the other is created The verification of these processing reports greatly reduces the risk of introducing errors in the data at the processing stage Verification Electronic text files containing the daily transmissions of completed cases are combined to create the raw survey file All EICS records could be matched to their corresponding record from the LFS
83. yment in this survey means that the persons usually work 30 hours or more per week in their job or jobs Part time employment consists of all other persons that is those who usually work less than 30 hours per week The LFS defines part time work differently for multiple job holders it applies the 30 hour criterion only to the main job Insurable employment Refers to work that is insured by the Employment Insurance El program against an interruption of earnings Self employment and some other types of employment are excluded The survey identifies insurable employment based on the person having EI premiums deducted from their pay and the class of worker El Claimant A claimant is a person who submitted an EI claim during a specified period El Beneficiary A beneficiary is someone who upon claiming El benefits qualifies and receives benefits for a particular period for instance the reference week the reference month or since the last work interruption Potentially eligible for El Term used in analysis to describe unemployed people who during the reference week received El benefits or were in a position to receive them because of their recent insurable employment and subsequent job loss This includes all unemployed persons with some insurable employment in the last 12 months who did not quit their job without cause or in order to return to school Eligible for El This is a subset of the potentially eligible population It inclu

Download Pdf Manuals

Related Search

Related Contents

設計から現場への図面の流れ（建築・土木・製造・保守）  MSRT-Pro 1-3K Neutral English Manual  NX650 Service manual index  Page 1 Page 2 前書き 本取扱説明書は、 携帯』_のバッテリ…装置を  Casio 5099 Watch User Manual    StarTech.com 6 ft DB25 to Centronics 36 Parallel Printer Cable - M/M    

Copyright © All rights reserved. Failed to retrieve file