Transcript Slide 1
Using the IEA IDB Analyzer Correlations & Regression ICCS 2009 IDB Workshop, 18th February 2010, Madrid Table of content Correlations – – – – Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Regression – – – – – Linear Regression Model Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Hands-On Training ICCS 2009 IDB Workshop, 18th February 2010, Madrid Table of content Correlations – – – – Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Regression – – – – – Linear Regression Model Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Hands-On Training ICCS 2009 IDB Workshop, 18th February 2010, Madrid Calculating Correlations Choose variable(s) to be analyzed Load the SPSS Analysis File Choose Correlations as analysis type (some settings will be done automatically with respect to the datafile and the analysis method chosen) Select Variables from the datafile to be analyzed as Analysis Variables Define location and name of the Output Files Start SPSS and run analysis ICCS 2009 IDB Workshop, 18th February 2010, Madrid Select Variable of Interest Student Questionnaire, Q14A-F, p. 13 Variables of Interest: PARTCOM PVCIV01-05 ICCS 2009 IDB Workshop, 18th February 2010, Madrid The Analysis File The analysis file needs to be loaded from the respective directory Countries in the example analysis file: – – – – – – – Austria Belgium (Flemish) Bulgaria Denmark England Estonia Finland ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Preparation Select Analysis File: C:\ICCS2009\Work\ICG_ISG_INTC2.sav Select Analysis Type: Correlations Select Analysis Variables: PARTCOM (Students’ civic participation in the wider community) Select Achievement Variables: PVICIV01-05 (Table 5.09 from ICCS2009 International report, first column) ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details Select Analysis File C:\ICCS2009\Work\ICG_ISG_INTC2.sav ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details 1) Select Correlations 2) Check Exclude Missing from Analysis 3) Check With Achievement Scores ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details IDCNTRY TOTWGTS JKZONES ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details Search for PARTCOM ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details Search results will be displayed ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details Highlight variable Use arrow key to add variable to analysis ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details Highlight variable Use arrow key to add variable to analysis ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details Change number of decimals to “1” Define path and filename for output: “C:\ICCS2009\Work\Table_5_09.*” ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Details The IDB Analyzer creates SPSS Syntax and starts SPSS In SPSS Syntax Editor Choose: Run > All ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations - Outcome As a result the IDB Analyzer creates the following in the working directory (C:\ICCS2009\Work\): – SPSS Syntax file – contains the syntax with the commands (*.sps) – SPSS Data file – contains statistics from the analysis (*.sav) – MS Excel Output file – contains statistics from the analysis (*.xls) ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations – SPSS Output ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations – Excel Output ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations – Excel Output List of Countries Achievement Scores Sum of Weights Mean Achievement S.E. of Mean Achievement ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations – Excel Output Standard Deviation of Mean Achievement Correlation of PV with itself S.E. of Standard Deviation of Mean Achievement ICCS 2009 IDB Workshop, 18th February 2010, Madrid S.E. of Correlation Correlations – Excel Output Variable Name of Second Variable S.E. of Mean of Variable Mean of Variable (here: Country Mean) S.E. of Standard Deviation of Mean of Variable Standard Deviation of Mean of Variable ICCS 2009 IDB Workshop, 18th February 2010, Madrid Correlations – Excel Output Correlation of variable 1 with variable 2 (here: Achievement with Participation in Community) ICCS 2009 IDB Workshop, 18th February 2010, Madrid S.E. of Correlation Correlations – Excel Output ICCS 2009 IDB Workshop, 18th February 2010, Madrid Table of content Correlations – – – – Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Regression – – – – – Linear Regression Model Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Hands-On Training ICCS 2009 IDB Workshop, 18th February 2010, Madrid Table of content Correlations – – – – Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Regression – – – – – Linear Regression Model Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Hands-On Training ICCS 2009 IDB Workshop, 18th February 2010, Madrid Linear Regression Model ICCS 2009 IDB Workshop, 18th February 2010, Madrid Linear Regression Model y = β0 + x β1 y is the dependent variable – here: estimated mean of all 5 plausible values x is the independent variable ß0 is the intercept (value of y when x is zero) ß1 is the slope (change in y for each unit increase in x) ICCS 2009 IDB Workshop, 18th February 2010, Madrid Computing Regression Can be used to calculate regression coefficients and their (jackknifed) standard errors for the ICCS background variables Uses the jackknifing procedure and therefore considers the sampling method used in ICCS Makes use of the variables JKZONE and JKREP ICCS 2009 IDB Workshop, 18th February 2010, Madrid Computing Regression Choose variable to be analyzed Load the SPSS Analysis File Choose Regression as analysis type (some settings will be done automatically) Select Variables from the datafile to be analyzed as Analysis Variable AND/OR Select the plausible values as Achievement Scores Define location and name of the Output Files Start SPSS and run analysis ICCS 2009 IDB Workshop, 18th February 2010, Madrid Recoding variables Some tables might display data differently than in the data files E.g., table 7.1 has been calculated using dummy recoded information of the students‘ immigration background (variable name in the datafile is IMMIG) For replication of the table, the information from the variable IMMIG needs to be recoded in SPSS and added as a new variable to the datafile before running the analysis ICCS 2009 IDB Workshop, 18th February 2010, Madrid Dummy Coding for Regression IMMIG Reg01IMMIG Native 1st Generation Immigrant Non-Native IMMIG 1 2 3 Reg01IMMIG 0 1 1 ICCS 2009 IDB Workshop, 18th February 2010, Madrid Dummy Coding for Regression IMMIG values: System Missing 0 1 1 ICCS 2009 IDB Workshop, 18th February 2010, Madrid SPSS: Dummy Coding for Regression Menu: TRANSFORM Recode into Different Variables... ICCS 2009 IDB Workshop, 18th February 2010, Madrid SPSS: Dummy Coding for Regression Menu: TRANSFORM Recode into Different Variables... 10 21 31 ELSE SYSMISS ICCS 2009 IDB Workshop, 18th February 2010, Madrid Linear Regression Model Predictor variable: Reg01IMMIG y = β + xβ 0 1 Mean achievement for native students Difference between mean achievement of native and mean achievement of non-native students ICCS 2009 IDB Workshop, 18th February 2010, Madrid The Analysis File The analysis file needs to be loaded from the respective directory Countries in the example analysis file: – – – – – – – Austria Belgium (Flemish) Bulgaria Denmark England Estonia Finland ICCS 2009 IDB Workshop, 18th February 2010, Madrid Computing Regression Select data file: C:\ICCS2009\Work\ICG_ISG_INTC2_Reg.sav Analysis Type: Regression Grouping Variable: IDCNTRY Analysis Variable: Reg01IMMIG Achievement Scores: PVICIV01-05 (Table 7.1, first column from ICCS2009 International report) ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression - Details C:\ICCS2009\Work\ICG_ISG_INTC2.sav 1) Select Regression 2) Check Exclude Missing from Analysis 3) Check With Achievement Scores IDCNTRY Reg01IMMIG PVICIV01-05 Change the number of decimals to 2 TOTWGTS JKZONES C:\ICCS2009\Work\Table_7_1.* ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression - Details The IDB Analyzer creates SPSS Syntax and starts SPSS In SPSS Syntax Editor Choose: Run > All ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression - Outcome As a result IDB Analyzer creates the following in the working directory (C:\ICCS2009\Work): – SPSS Syntax file – contains the syntax with the commands – SPSS Data file – contains statistics from the analysis – MS Excel Output file – contains statistics from the analysis ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression – SPSS Output Number of Cases Multiple R-Squared S.E. of Intercept Intercept: Mean Achievement of Native Students ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression – SPSS Output Estimate of Regression Coefficient S.E. of Regression Coefficient T-Statistics of Regression Coefficient ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression – SPSS Output ABS (-1.3) < 1.96 In Bulgaria the difference in civic knowledge between native students and non-native students is NOT significant. ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression – SPSS Output ABS (-8.9) > 1.96 In Austria the difference in civic knowledge between native students and non-native students IS significant. ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression – SPSS Output Differences in Draft Table 7.1 of Int. Report are incorrect ICCS 2009 IDB Workshop, 18th February 2010, Madrid Regression – SPSS Output Predictor variable: Reg01IMMIG y = 516,2 - x 56,6 Mean achievement for native students Difference between mean achievement of native and mean achievement of non-native students ICCS 2009 IDB Workshop, 18th February 2010, Madrid Thank you for your attention! Any Questions? ICCS 2009 IDB Workshop, 18th February 2010, Madrid Table of content Correlations – – – – Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Regression – – – – – Linear regression Model Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Hands-On Training ICCS 2009 IDB Workshop, 18th February 2010, Madrid Table of content Correlations – – – – Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Regression – – – – – Linear regression Model Settings for the Analysis Running the Analysis Compare SPSS Output and International Report Output in MS Excel Hands-On Training ICCS 2009 IDB Workshop, 18th February 2010, Madrid Hands-On Training A. Re-produce the example using your country data - Correlation of students’ participation in the wider community with civic and citizenship achievement (PVCIV01-05 with PARTCOM) and/or B. Re-produce the example using your country data - Regression of students’ immigration status on civic and citizenship achievement of (IMMIG [recoded] on PVCIV01-05) and/or C. Practice with own selected variables following these analysis steps ICCS 2009 IDB Workshop, 18th February 2010, Madrid