GeSCA User's Manual
Contents
1. P Enlarge Size Reduce Size Delete One Delete All lt Uploaded data Internet R 100 Once the above steps are complete users can run GeSCA for fitting the specified model to multiple groups simultaneously This is done by clicking on the RUN icon on the top of the program As shown below all multi group analysis results are displayed in the Results window In this example the same model was applied to two groups males and females at the same time 20 Thus all parameter estimates are provided for each of the two groups labeled Group 1 and Group 2 GeSCA Windows Internet Explorer le http fwww sem gesca com gsca3 php Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis A aA 2 E Home Save Model Options Indicators Results E Indicators Measurement Model Done Internet 100 21 2 How to conduct a multi group analysis with cross group equality constraints e To impose cross group equality constraints on loadings select indicators whose loadings are constrained to be equal across groups in the Assign Indicators window Then constrain the loadings of the selected indicators to be identical across groups by inserting a label in the Cross Group Equality Constrai2. LS Draw Latent Variables es Move Latent Variables Assign Indicators Draw Path Coefficients Enlarge Size i Reduce Size gt Delete One gt Delete All Uploaded data 1 3 3 3 4 3 1 4 5 3 4 4 a E E E WW O o now YN A JNA O fe J w e JN Ww A a Q Q a W J W w J W w il e Click the left mouse button with the cursor placed in the Model Specification window of the program as many times as the number of latent variables In the present example four clicks resulted in the creation of four latent variables By default the four latent variables were initially named LV_ to LV_4 GeSCA Windows Internet Explorer e http fiwaw sem gesca com gsca3 php Total o item is waiting on the server Total o item is executing File View Analysis p EE J J la gt aa 2 Home Upload Data Open Model RUN Save Model Option Indicators c3 Indicators i Draw Latent Variables he Co Unused Indi C Used Indica O eed 1 es Move Latent Variables CILVv_2 GILv_3 ga Assign Indicators gt Draw Path Coefficients P Enlarge Size Reduce Size h a Delete One gt Delete All lt ji 1 Uploaded data 1 3 3 3 4 3 3 3 3 4 4 4 4 3 4 3 2 3 4 3 3 3 fast
3. Je je http jjwww sem gesca comjgsca3 php v A Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis F gt A ia mm gt Home Upload Data Open Model Save Model Indicators Model Specification a Indicators C Unused Indi C Used Indicaj es Move Matent Variables sa Assign Indicators Draw Path Coefficients Z5 Options atent Variables LY orgen PN Enlarge Size S Reduce Size gt Delete One gt Delete All i i gt i Uploaded data internet R 100 15 In the Constrain Path Coefficients window constrain the selected path coefficient by inserting a label in the Equality Constraints dialog Then click on OR GeSCA Windows Internet Explorer http www sem gesca com gscas php Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis Options Home Upload Dats Open Model RUN Save Model Indicators C3 Indicators C Unused Indi C Used Indicaj Constrain Path Coefficients gt No Constraints Equality Constraints T i User Defined Constraints OK CANCEL Eo Uploaded data LS Draw Latent Variables es Move Latent Variables Assign Indicators gt
4. gt GeSCA Windows Internet Explorer TA w 2 http fiwaw sem gesca com gsca3 php A otal o item is waiting on the server Total o item is executing z few Analysis R la Z gt 2 2 Home Upload Data Open Model RUN Sawe Model Options CEEA z Model Specification C Used Indica gt c Lv 2 4 ea I Assign Indicators Cy orge Ci orge gt Draw Path Coefficients pa sae Cw ce A Enlarge Size amp Reduce Size 4 i Uploaded data Internet 100 2 To view the results of the previous analysis choose View and then Results on the very top menu f GeSCA Windows Internet Explorer 2 http iwww sem gesca com gsca3 php otal o item is waiting on the server Total o item is executing Results is gt By H Options Upload Data Open Model RUN Save Model Model Specification LS Draw Latent Variables es Move Latent Variables Assign Indicators CA s4 Draw Path Coefficients GA Enlarge Size a Reduce Size Qv Internet 25
5. gt cI Unused Indi L5 Draw Latent Variables C Used Indica O eed L141 es Move Latent Variables G Lv_2 4 ea Assign Indicators 5 orge rs orge Ga Draw Path Coefficients C orger PE Enlarge Size Reduce Size gt Delete One gt Delete All 4 ji Uploaded data Internet R 100 As shown below all analysis results are displayed in the Results window Dar le http www sem gesca com gsca3 php Rc a Pa GeSCA Windows Internet Explorer Your analysis is being conducted Total o item is waiting on the server Total 1 item is executing File View Analysis ay Home Indicators 4 Indicators e Draw Latent Variables C Unused Indi C Used Indica gt ied LV 1 es Move Latent Variables cI LY 2 ii me Assign Indicators CY orge CY orgen gt Draw Path Coefficients CS orgen PN Enlarge Size O Reduce Size AVE 0 641 Alpha 0 919 Y Delete One a Delete All 0 161 0 006 364 0 679 0 030 ia Uploaded data R 100 Users can copy and paste the results into WORD by using shortcut functions In the Results window Ctrl a Select all Ctrl c Copy Ctrl v Paste HOW TO INTERPRET GeSCA RESULTS Model Fit 0 535 0 532 0 993 0 078 This table provides four measures of
6. 1 4 5 3 4 4 3 3 3 2 2 3 3 2 3 3 4 3 3 3 3 a Z E a E 7 a ei a 4 e FJ lt gt Dane ED Internet inne X Step 2 Assign indicators to latent variables measurement model After drawing latent variables users are to specify their measurement model as follows e Click once on the Assign Indicators icon in the right hand side of the program and then click on an individual latent variable a circle on the Model Specification window GeSCA Windows Internet Explorer http www sem gesca com gsca3 php Total o item is waiting on the server Total o item is executing File View Analysis AD ia gt A Home Upload Data Open Model Save Model Indicators C3 Indicators gt C Unused Indi cC Used Indica lt M Uploaded data Model Specification Ce Variables atent Variables Assign Indicators gt Draw Path Coefficients si Enlarge Size Reduce Size gt Delete One gt Delete All J Internet 100 e The latent variable can be renamed by typing its label in the Assign Indicators window e Select the appropriate indicators in the list which appears on the left hand dialog window and move them to the right hand dialog window Free means a free loading to be estimated e Choose whether the selected indicators are to be specified as Refl
7. 2 9 go 6 E f a Fa Tg eg p l 3 J 4 4 J i 2O 2 g 4 amp 4 amp g f 2 et 4 Se a j i j j J j 3 Open GeSCA When users open GeSCA for the first time they will see the following warning sign Click on Run in order to start the program _GESCA Windows Internet Explorer le http qsca ath cx gsca php Total o item is waiting on the server Total o item is executi Warning Security The application s digital signat ce cannot be verified Do you want to run the applicati Name newpackage New JApplet Publisher Heungsun Hwang Ph D http fgsca ath cx Cancel WY The digital signature cannot be verified by a trusted source 4 vore Information run if you trust the origin of the application ws amp internet R100 v 4 Upload data Users can upload their data file by clicking on the icon Upload Data on the top menu of the program r GeSCA Windows Internet Explorer http waw sem gesca com gsca3 php Total o item is waiti tal o item is executing File View Analysis F i ie gt Home Upload Data Open Model RUN Save Model Indicators odel Specification Indicators d Draw Latent Variables gt c Unused Indi gt c Used Indica Assign Indicators es Move Latent Variables lt Draw Path Coefficients i GX Enlarge Size eR Reduce Size gt Delete One gt D
8. 13 76 0 045 15 09 12 55 0 039 16 98 ma 0 743 0 040 18 44 0 184 0 021 8 64 0 552 0 060 9 27 i LV_3 AVE 0 589 Alpha 0 766 16 41 0 053 21 08 0 038 orgemt3 19 14 0 033 0 033 16 52 0 047 10 5 16 64 20 34 10 72 AVE 0 583 Alpha 0 641 0 029 18 4 0 634 0 046 13 83 15 08 0 070 7 2 15 21 0 047 12 91 CR significant at 05 level This table displays the estimates of loadings and weights of individual indicators The SMC Squared Multiple Correlation of each indicator is equivalent to its squared loading indicating how much variance of an indicator is explained by the corresponding latent variable The absolute bootstrap critical ratio CR is obtained by dividing a parameter estimate by its bootstrap standard error SE The CR is used for testing the significance of an estimate For example an estimate may be considered significant at 05 level 1f its bootstrap critical ratio is equal to or larger than two in absolute value under the assumption that the bootstrap distribution of the estimate is roughly normal The AVE Average Variance Extracted is the average amount of variance of indicators that is explained by their corresponding latent variable The Alpha indicates Cronbach s alpha When indicators are formative their loadings and SMCs will not be reported Structural Model Path Coefficients Estimate SE CR 0 067 0 042 0 049 CR si
9. 3 3 4 4 4 5 4 4 4 10 1 4 4 4 4 4 4 4 4 4 4 4 5 4 4 11 1 4 5 3 5 5 3 3 4 3 4 5 4 4 4 lis 3 1 5 4 4 5 5 4 4 5 4 4 4 4 4 4 EEX 1 4 4 4 4 4 4 4 4 3 3 3 3 3 3 14 1 4 4 4 5 5 5 5 5 4 4 3 4 3 4 as 1 4 4 3 4 4 3 4 4 4 4 4 3 3 4 16 1 4 4 3 4 5 5 4 5 4 4 4 4 3 4 232 1 2 2 3 3 3 2 3 3 3 3 2 3 5 4 18 1 4 4 3 4 4 4 4 4 4 4 4 3 4 4 jett 1 5 4 4 4 4 4 4 4 4 3 4 3 4 4 20 1 4 4 4 4 4 4 4 4 4 4 4 4 4 5 21 1 4 4 3 4 4 4 3 4 4 4 4 4 4 4 22 1 5 4 4 4 5 4 3 4 4 4 5 4 5 4 23 1 5 5 4 5 5 5 5 5 2 3 5 4 3 3 24 1 1 3 3 2 1 2 2 2 3 3 4 3 3 4 25 1 5 5 5 5 5 5 5 5 3 3 5 5 5 4 26 1 5 4 4 4 3 3 3 4 4 4 5 4 4 4 az i c c cl cl al a al a a a a gt a a z 4 gt bt rick2_O lt gt Alternately the raw data file can be prepared in ASCH format txt or dat The specific data format required for this case is as follows e The first row contains the names of indicators The name of each indicator should be separated by a space e The data input begins on the second row Data from an observation or responses by an individual on each indicator should be separated by a space e Data for each observation should appear on a single row The following shows the first five rows of the data set created in ASCII format for the present example ceil cei cel cel4 cei celb cel cei mal mad mas mad mas mab orgentl orgemte orgemts orgemt orgemts orgemt6 orgente E 2a 2 f amp E esate 2 fF F amp F FF wb Gg 4 z 4 So d T E a A d dad d dadd 4 1 J j j i
10. 3 and 7 and Affective Commitment Love AC_ Love by three indicators orgemt5 6 and 8 ceil cei2 orgcmt1 orgcmt2 cei3 crgcmt7 cei5 cei6 orgemt5 Org_lIden cei7 orgcmt6 orgcmt8 amp cei8 mat ma2 ma3 ma4 mad ma6 Figure 1 The specified structural equation model for the example data 2 Prepare a raw date file GeSCA is run on individual level raw data The raw data file is to be prepared in Excel xls or xlsx The specific data format required is as follows e The first row contains the names of indicators e The data input begins on the second row The following shows the data set created in Excel for the present example EJ Microsoft Excel rick_gsca xls ieJ File Edit view Insert Format Tools Data Window Help Adobe PDF Type a questionforhelp X SF ed oh Ry Calibri ij A1 y amp gender E o c D E F G H J K L M N o FI A a gender iceil cei2 cei3 ceig cei5 cei cei ceig mal ma2 ma3 ma4 ma5 ma org 2 1 3 3 3 4 3 3 3 3 4 4 4 4 3 4 IES 1 4 5 3 4 4 3 3 3 2 2 3 3 2 3 4 5 5 4 4 5 4 4 5 3 2 4 3 2 4 ES 1 4 4 4 4 4 4 4 4 4 4 4 4 4 4 6 a 4 4 4 4 4 4 4 4 4 4 4 4 4 4 Eza 1 4 4 4 4 4 4 4 4 3 3 4 3 3 2 8 1 4 4 3 5 5 3 3 4 3 4 4 3 4 2 9 1 4 3 3 4 4
11. Draw Path Coefficients A Enlarge Size i Reduce Size Delete One ho Delete All P a ee Internet A 100 Repeat the above step for other path coefficients that are to be held equal to the first path coefficient using the same label In the Model Specification window subsequently users can see all chosen paths labeled the same b indicating that they are constrained to be equal to each other f GeSCA Windows Internet Explorer http waw sem gesca com gsca3 php Your analysis is now complete Total o item is waiting on the server Total o item is executing l E By a X Save Model File View Analysis la Home Upload Data Open Model Options Indicators Model Specification 5 Indicators l Draw Latent Variables C Unused Indi C Used Indica O m L141 es Move Latent Variables m Lv 2 c Ly 3 x cI Lv 4 Assign Indicators C orger Cy orge gt Draw Path Coefficients Cy orger GQ Enlarge Size Reduce Size gt Delete One gt Delete All M rl Uploaded data 1 3 4 3 4 3 3 3 3 4 4 4 4 3 4 a 2 3 4 3 3 3 at 1 4 5 3 4 4 a 3 3 2 2 fa 3 2 3 1 3 3 3 3 3 as E E e a l n l I a i rl i M lt gt Done Internet A i100 16 5 How to handle second order latent variables We de ce
12. Variables gt cJ Lv_2 LY_3 e es Move Latent Variables Ci orge G orgen s D orge C C EA Assign Indicators gt EI Lv_4 E g Draw Path Coefficients PN Enlarge Size Options Upload Data Open Model RUN Sawe Model el Reduce Size hN Delete One gt Delete All il no Uploaded data 4 Users draw path coefficients to complete their structural model as described on page 6 18 MULTI GROUP ANALYSIS Note To conduct a multi group analysis users must include a categorical grouping variable in the data which indicates group memberships of cases e g gender 1 male amp 2 female 1 How to conduct a multi group analysis without cross group equality constraints e To begin select Multi Group Analysis under the top menu of Analysis GeSCA Windows Internet Explorer a item is executing gt Indicator Model Specification C3 india Draw Latent Variables C Unused Indi LY Empty gt c Used Indica gt Move Latent Variables Assign Indicators gt Draw Path Coefficients Enlarge Size Reduce Size gt Delete One SS Delete All ji gt i Uploaded data Internet e Upload a data file by clicking on the icon Upload D
13. Last updated November 28 2011 Heungsun Hwang GeSCA User s Manual Content Page HOW TO USE GeSCA 1 Sample data and model l 2 Prepare a raw date file l 3 Open GeSCA 2 4 Upload data 3 5 Specify a structural equation model 4 6 Run GeSCA 7 HOW TO INTERPRET GeSCA RESULTS 8 OTHER OPTIONS IN SINGLE GROUP ANALYSIS 1 How to constrain loadings to user defined values 11 2 How to impose equality constraints on loadings 12 3 How to impose a user defined constraint on path coefficients 13 4 How to impose equality constraints on path coefficients 14 5 How to handle second order latent variables 16 MULTI GROUP ANALYSIS 1 How to conduct a multi group analysis without cross group equality constraints 18 2 How to conduct a multi group analysis with cross group equality constraints 21 MISSING DATA 23 OTHER OPTIONS 24 HOW TO USE GeSCA 1 Sample data and model Bergami and Bagozzi s 2000 organizational identification data are used for illustrative purposes The number of observations is equal to 305 The model specified for the data is displayed in Figure 1 No residual terms are displayed in the figure As shown in the figure this model consists of four latent variables and 21 reflective indicators Specifically Organizational Prestige Org Pres is measured by eight indicators ceil cei8 Organizational Identification Org Iden by six indicators mal ma6 Affective Commitment Joy AC_ Joy by four indicators orgcmt1 2
14. Reduce Size h S Delete One Loom Delete All vw gt Internet A 100 e Drag a path from an exogenous latent variable to the corresponding endogenous latent variable Repeat the above steps until all paths are drawn f GeSCA Windows Internet Explorer DaS http jweaw sem gesca com gsca3 php Total o item is waiting on the server Total o item is executing File View Analysis A a n A Le aun Upload Data Open Model Save Model Indicators Model Specification C5 Indicators gt i gt c Unused Indi Draw Latent Variables C Used Indica gt GIL 7 move Latent variables G Lv_2 G ILv_3 Assign Indicators gt Draw Path Coefficients i PE Enlarge Size i Reduce Size Delete One gt Delete All lt il Uploaded data Internet R 100 6 Run GeSCA Once the above steps are complete users can run GeSCA for fitting the specified model to the data This is done by clicking on the RUN icon on the top of the program GeSCA Windows Internet Explorer TAS v http waw sem gesca com gsca3 php Total o item is waiting on the server Total o item is executing File View Analysis is gt By s weet ITTY Home Upload Data Open Model RUN Save Model Indicators Model Specification Indicators
15. V_3 x Go A o Equality Constraints Cy orge Assign Indicators Ci orge A User Defined Constraints ry orger LS Draw Latent Variables gt Move Latent Variables gt Draw Path Coefficients 4 Enlarge Size e _ Reduce Size gt Delete One a Delete All lt i Uploaded data 1 es Internet 14 In the Model Specification window subsequently users can see the path fixed to the defined value Seles http fwww sem gesca com gsca3 php v A GeSCA Windows Internet Explorer Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis A 9 mS x Upload Data Open Model Save Model Options Indicators Model Specification E Indicators Draw Latent Variables gt C Unused Indi C Used Indica es Move Latent Variables Assign indicators gt Draw Path Coefficients CY orgen a Enlarge Size eR Reduce Size Delete One sy Delete All Internet R 100 4 How to impose equality constraints on path coefficients e Click once on the Assign Indicators icon in the right hand of the program and then click on an individual path to be constrained in the Model Specification window gt GeSCA Windows Internet Explorer
16. ata Then users are to select their grouping variable Then click on OK In this example gender was chosen 19 GeSCA Windows Internet Explorer fey w http jjwww sem gesca comjgsca3 php Total o item is waiting on the server Total o item is executing Upload Data Open Model Save Model Options Indicators CJ Indicators LS Draw Latent Variables es Move Latent Variables cei2 cei3 z S gt Assign Indicators cei4 cei5 cei6 cei7 ceig ma1 ma2 lt A Enlarge Size ma3 lt Draw Path Coefficients Reduce Size h Delete One gt Delete All OK Cancel v gt Internet R ioo Users then specify their measurement and structural models in the Model Specification window in the same way as described earlier see pages 4 to 6 GeSCA Windows Internet Explorer T E x lt http weaw sem gesca com gsca3 php bl AA Total o item is waiting on the server Total o item is executing File View Analysis i F iF Home Upload Data Open Model RUN Save Model Options Indicators Model Specification ca Indicators j p gt c Unused Indi Draw Latent Variables cC Used Indica O ed L141 gt Move Latent Variables G Lv_2 pea Assign Indicators Ci orge Ly orge gt Draw Path Coefficients C orgen
17. e lLv_1 Reflective ox O Formative CANCEL In the Assign Indicators window select an indicator whose loading 1s to be fixed to a Save Model ri seer x Options Constraints No Constraints D Equality Constraints User Defined Constraints on _ EAr A Draw Latent Variables es Move Latent Variables Assign Indicators gt Draw Path Coefficients a Enlarge Size Reduce Size 3 3 4 5 E E Qv Internet Constrain the loading of the selected indicator to a user defined value by inserting that value in the User Defined Constraints dialog box Then click on OKR r A GeSCA Windows Internet Explore a http fwwiw sem gesca com gsca3 php EDER A File View Analysis ia gt By 2 Home Upload Data Open Model RUN Save Model Options Indicators Assign Indicators i Constraint E Indicators simi LD Draw Latent Variables cI Unused Indi Sele All gt No Constraints Used Indica gt ceil Free Font x i EM 4 zi cei2 Free Equality Constraints es Move Latent Variables ee cei3 Free o lt lt z y cei4 Free i i mm L4 cei5 Free User Defined Const
18. ective default or Formative GeSCA Windows Internet Explorer 2 http sfwww sem gesca com gscas php Total o item is waiting on the server Total o item is executing File View Analysis A 5 ra ie sim Home a Save Model Indicators Indicators p Draw Latent Variables o Gand Unused Indi Select All Free Free Free Free Assign Indicators Free Free Fres gt Draw Path Coefficients Free es Move Latent Variables ba Enlarge Size Name L_1 Reflectwe Formative Sp Pee Sire OK CANCEL h a Delete One w Delete All M Uploaded data g gt Internet R 100 e Kepeat the above steps tor the remaining latent variables Step 3 Draw path coefficients structural model Path coefficients are to be drawn as follows e Click once on the Draw Path Coefficients icon in the right hand side of the program GeSCA Windows Internet Explorer 2B http jjwww sem gesca comigsca3 php x Total o item is waiting on the server Total o item is executing File View Analysis Home Upload Data Open Model Save Model Indicators Model Specification Ci Indicators C Unused Indi Ey Used Indica LD Draw Lateng Variables agent Variables Indicators gt Draw Path Coefficients s Enlarge Size i
19. elete All i gt i Uploaded data Done D Internet A 100 v Subsequently the list of indicator names in the input data file will appear in the left window of the program under the label Indicators ra f GeSCA Windows Internet Explorer E http waw sem gesca com gsca3 php Total o item is waiting on the server Total o item is executing File View Analysis A a gt mS x Home pload Dats Open Model RUN Save Model Options Indicators Model Specification cI Indicators LS Draw Latent Variables es Move Latent Variables Assign Indicators gt Draw Path Coefficients PE Enlarge Size Reduce Size Delete One hos Delete All Internet R 100 5 Specify a structural equation model Users can specify their structural equation model with the following steps Step 1 Draw latent variables Users are to draw latent variables before assigning indicators to them as follows e Click once on the Draw Latent Variables icon in the right hand side of the program Bees GeSCA Windows Internet Explorer http fwww sem gesca com gscas php A Total o item is waiting on the server Total o item is executing File View Analysis Home pload Data Open Model Indicators Model Specification C3 Indicators RUN x Save Model Options
20. gnificant at 05 level This table shows the estimates of path coefficients and their bootstrap standard errors SE and critical ratios CR 10 R square of Latent Variable S 9 S O 0 131 0 377 0 163 This table provides the R square value of each endogenous latent variable indicating how much variance of an endogenous latent variable is explained by its exogenous latent variables In the present example the first latent variable LV_1 is exogenous only Thus its R square is equal to zero This table provides the mean score estimate of each latent variable va 1 2 y3 V4 V l 0 388 0 055 0 614 0 042 0 461 0 0467 0 209 0 060 0 404 0 049 0 461 0 046 significant at 05 level This table shows the correlations among latent variables along with their bootstrap standard errors in parenthesis 11 OTHER OPTIONS IN SINGLE GROUP ANALYSIS 1 How to constrain loadings to user defined values user defined value er i http weaw sem gesca com gsca3 php File View Analysis Indicators J Indicators C Unused Indi n ETAR m3 Lv_2 ecw eI Ly_4 Cy orgen i orge C orge C Used Indicaj lt il EEJ Uploaded data GeSCA Windows Internet Explorer Upload Data le Open Model RUN IE Assign Indicators ma4t mas ma6 orgcmtt orgcemt2 orgcmts orgcemt7 orgcmt5 orgcmt6 orgcmts Nam
21. n the Assign Indicators icon in the right hand of the program and then click on an individual path to be constrained in the Model Specification window GeSCA Windows Internet Explorer BA 2 http fwwiw sem gesca com gscas php w Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis A E Home Upload Data A Open Model 3 Save Model Model Specification Indicators a oRERD S atent Variables gt cC Unused Indi CI Used Indicaj gt c va m E Lbs Av Cy orge Ci orge Ly orgen es Move Matent Variables gt Draw Path Coefficients s Enlarge Size Reduce Size gt Delete One gt Delete All internet A 100 In the Constrain Path Coefficients window constrain the selected path coefficient to a user defined value by inserting that value in the User Defined Constraints dialog Then click on OR GeSCA Windows Internet Explorer 2 http fwwiw sem gesca com gsca3 php Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis 2 LS Lee a Open Model Save Model Indicators Constrain Path Coefficients Indicators gt c Unused Indi C Used Indica f o GILVv_1 CO No Constraints o Ga A gt aad L
22. nts dialog box Then click on OK 2 GeSCA Windows Internet Explorer EBS http www sem gesca com gsca3 php S A Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis A BS gt 2 Home Upload Data Open Model RUN Save Model Options Indicators Assign Indicators em Indicators Constraints Draw Latent Variables amp EI Unused Indi eit O No Constraints Cd Used Indica ma4 Fa F j TE gt gt ceil f i gt GILV_1 ma5 cei2 Cross Group Equality warts nts es Move Latent Variables gt cI Lv 2 ma6 cei3 a O ad LV_3 orgemt1 lt lt i cei4 2 3 ma L4 orgcmt2 cei5 Free O User Defined Cross Group Constraints AEE orge orgemt3 cei6 Free Ci orge jorgemt7 cei7 Free Draw Path Coefficients C orge jorgcmts cei8 Free orgemt6 OK orgcmt x i Enlarge Size Name Lva 5 A Reduce Size Reflective O Formative 3 OK CANCEL gt Delete One gt Delete All oy gt i Uploaded data 1 amp 38 amp 2 amp 2 amp 2 amp 3 92 amp 2 amp 2 S amp S a Ss a SB E 1 4 5 3 4 4 3 3 2 2 3 3 2 3 3 1 3 3 3 3 3 4 E E A A Cas A A E 3 9 A 3 A 4 3 3 4 3 93 lt gt Done Internet A 100 e To impose cross group equality constraints on pa
23. overall model fit and the number of free parameters FIT indicates the total variance ofall variables explained by a particular model specification The values of FIT range from 0 to 1 The larger this value the more variance in the variables is accounted for by the specified model AFIT Adjusted FIT is similar to FIT but takes model complexity into account The AFIT may be used for model comparison The model with the largest AFIT value may be chosen among competing models Unweighted least squares GFI and SRMR standardized root mean square residual Both are proportional to the difference between the sample covariances and the covariances reproduced by the parameter estimates of generalized structured component analysis The GFI values close to 1 and the SRMR values close to 0 may be taken as indicative of good fit For example SRMR lt 08 is indicative of an acceptable model fit NPAR is the number of free parameters estimated including weights loadings and path coefficients Measurement Model __ Estimate SE CR Estimate SE Weight SE CR eee AVE 0 641 Alpha 0 919 0 010 14 5 0 609 _ 0 039 15 55 15 72 20 2 16 0 0 593 0 044 13 53 15 58 0 048 13 43 0 011 14 93 0 045 14 31 19 68 0 046 15 61 16 96 0 038 15 99 18 0 0 642 0 053 12 2 ee AVE 0 581 Alpha 0 855 0 021 10 57 14 95 10 77 0 038 15 01 11 54 0 050 8 12 mad 0 823 0 028 29 86 0 261 _ 0 019
24. raints T O orge cei6 Free Ci orge cei7 Free gt Draw Path Coefficients gt orge ceig Free A Enlarge Size Name Lv 1 ee z i Reduce Size Reflective 1 Formative CANCEL ha Delete One ho Delete All aw rie Uploaded data 1 3 3 3 3 3 3 3 4 4 4 4 3 4 2 3 4 3 3 x as 1 4 5 3 4 3 3 3 2 2 3 3 2 3 1 3 3 3 3 3 4 i E A E A A E I a el a 3 a a 3 A iv 12 2 How to impose equality constraints on loadings e Inthe Assign Indicators window select indicators whose loadings are constrained to be equal pe GeSCA Windows Internet Explorer http fwww sem gesca com gscas php File View Analysis gt N A IB om 25 Home Upload Data Open Model Save Model Options Indicators Assign Indicators Fe Indicators muy Constraints o cA Unused Indi 2 D No Constraints cC Used Indica LS Draw Latent Variables Equality Constraints es Move Latent Variables User Defined Constraints Assign indicators i orge gt Draw Path Coefficients mail GA Enlarge Size Name Lv_1 EX i eN Reduce Size Reflective I Formative Sa Sy Delete one Delete All 4 ll Uploaded data a Internet e Constrain the loadings of the selected indicators to be identical by inserting a label e g an alphabet letter or number in the Equali
25. sca First users specify first order latent variables as described on page 4 GeSCA Windows Internet Explorer EY v B http fwww sem gesca com gqsca php Es gt lt Pi File Edit View Favorites Tools Help E z gt gt w wa cesca fn D dh Page if Tools Total o item is waiting on the server Total o item is executing File View Analysis Home Upload Data Indicators Model Specification c Indicators Draw Latent Variables PS Uncced inal draw Latent Variabies C Used Indica GIL Draw 2nd Order Latent Variables GILv_2 LY _3 Saal gt Move Latent Variables B orgcy Cy orgen Cy orgcny C C A Assign Indicators o GILv_4 C gt Draw Path Coefficients Ga Enlarge Size Open Model RUN Save Model Options ei Reduce Size gt Delete One gt Delete All em gt i Uploaded data Subsequently users specify second order latent variables as follows 1 Click once on the Draw 2 Order Latent Variables icon in the right hand side of the program Click the left mouse button with the cursor placed in the Model Specification window of the program as many times as the number of second order latent variables which appear as blue circles In the example below a second order latent variable labeled LV2_5 is ass
26. tes Tools Help a ahr Gesca i Total o item is waiting on the server Total o item is executing File View Analysis f e P A B gt A 2x Home Upload Data Open Model RUN Save Model Options Indicators Assign Indicators Indicators LS Draw Latent Variables E Unused Ind C Used Indica GILv_1 Draw 2nd Order Latent Variables gt cJ Ly_2 LY_3 Ga LC orac es Move Latent Variables C4 orge C orge Assign Indicators gt gc Ly_4 gt Draw Path Coefficients Name Lv2_5 oe 2 E bJ Enlarge Size Reflective gt Formative gt Ok CANCEL Reduce Size Delete One gt Delete All l l gt i Uploaded data r m SS lt gt 3 Once the above step is complete paths connecting second order latent variables to their first order latent variables appear on Model Specification window 2 GeSCA Windows Internet Explorer Go E http ijwww sem gesca com gsca php E lt soo ked Fie Edit view Favorites Tools Help We te cesca a D A ma gt a Page iG Tools Total o item is waiting on the server Total o item is executing File View Analysis Indicators Model Specification J Indicators Draw Latent Variables G Unused Indi C Used Indica om es e Draw 2nd Order Latent
27. th coefficients select a path coefficient to be constrained by clicking once on the Assign Indicators icon and then clicking on its path in the Model Specification window Then constrain the selected path coefficient to be identical across groups by inserting a label in the Cross Group Equality Constraints dialog box Then click on OKR 22 r GeSCA Windows Internet Explorer Sele le http www sem gesca comj gsca3 php w Your analysis is now complete Total o item is waiting on the server Total o item is executing File View Analysis F D gt L p A ir a gt x Home Upload Data Open Model RUN Save Model Options Constrain Path Coefficients Indicators c Indicators a 2 Draw Latent Variables C Unused Indi 2 Cd Used Indica i GILv_1 _ No Constraints es Move Latent Variables GILV_2 GILV_3 EI Lv_4 Cross Group Equality Constraints Assign indicators TY orge LY orge gt Draw Path Coefficients ry orge ok CANCEL Enlarge Size Reduce Size Delete One ae Delete All im Uploaded data 1 3 3 3 4 3 3 3 s 4 4 4 4 3 4 3 2 3 4 3 3 3 5 1 4 5 3 4 4 z 3 3 2 2 3 2 4 3 1 3 3 3 3 3 4 E E A A E A 4 E 3 97 A r z A 4 3 2 A A 3 s5 gt O O RA D ino lt Note that any loadings and path coefficients with the same label will be constrained to be eq
28. ty Constraints dialog box Then click on OK GeSCA Windows Internet Explorer EBS le http www sem gesca com gsca3 php v A Your analysis is now complete Total o item is waiting on the server Total item is executing E File View Analysis j De a gt mix Home Upload Data Open Model Save Model Options Indicators Assign Indicators Constraints l J Indicators i e Draw Latent Variables J Select All ar C Unused Indi gt No ConsaQfaints Be C Used Indica maa 7 cei1 e 7 z mE ers 2z ne Equality Constraints es Move Latent Variables o Wa ma6 cei3 a PELs forgemes lt lt cei g cei4 Assign Indicators GILv_4 orgcemt2 eis User Defined Constraints O a orgemt3 cei6 LY orge orgcemt7 cei7 gt Draw Path Coefficients C orgey jorgemt5 ceig orgcmte PR Enlarge Size Name Lv 1 SN Reduce Size Reflective O Formative CANCEL Delete One Se Delete All i gt Uploaded data Internet R 100 13 Note that any loadings with the same label will be constrained to be equal to each other In the above example three indicators were chosen and labeled by an alphabet letter a This indicates that the loadings of the three indicators are to be held equal to one another 3 How to impose a user defined constraint on path coefficients e Click once o
29. ual to each other across groups 23 MISSING DATA GeSCA currently provides users with three options for dealing with missing observations 1 Listwise deletion 2 mean substitution and 3 least squares imputation To apply one of these options users choose the Options icon on the top of the program or Options under the menu of View Then select one of the options in the box of Missing Data The default value indicating a missing observation is 9999 However it can be changed to a user defined value gt GeSCA Windows Internet Explorer http www sem gesca com gsca3 php somplete Total o item is waiting on the server Total o item is executing A ii Home la li Results H amp Upload Data 2 Options j Open Model gt RUN Indicators Model Specification C3 Indicators C Unused Indi Cd Used Indica CS orgen CS orger CS orgen amp Save Model PS Options EEr A Estimation Options Iteration Maximum Number of Iterations Default O User defined Bootstrap Number of Samples Default O User defined Missing Data Missing Values Default O User defined Options Listwise deletion O Mean substitution O Least squares imputation Internet R 100 24 OTHER OPTIONS 1 To see the specified model again click on the Home icon to return to the Model Specification window
30. umed to underlie LV 3 and LV 4 gt GeSCA Windows Internet Explorer BEE J yy B http www sem gesca comigsca php all ine alee S Pir File Edit Yiew Favorites Tools Help BH dh Page Tools nf Total o item is waiting on the server Total o item is executing File View Analysis gS gt a um P d Home Upload Data Open Model RUN Sawe Model Options Indicators Model Specification 4 Indicators LS Draw Latent Variab C Unused Indi C Used Indica gt c Lva Draw 2nd Order Latent Variables gt cJ LVv_2 L _3 Ew gt Move Latent Variables Ci orgen Ci orgen 3 D orac D C3 seston indicators o GI Lv_4 C gt Draw Path Coefficients GQ Enlarge Size eX Reduce Size Delete One gt Delete All Diy j gt Uploaded data r r SS lt 17 2 After drawing second order latent variables click once on the Assign Indicators icon in the right hand side of the program and then click on an individual latent variable a circle on the Model Specification window Select the appropriate first order latent variables in the list which appears on the left hand dialog window and move them to the right hand dialog window T GeSCA Windows Internet Explorer oS ey 7 http jjv sem gesca comjgsca php File Edit View Favori
Download Pdf Manuals
Related Search