R
RCAAP Rss Feeder
Guest
Breve resumo:
Info Adicional:
Autor:
Clica para continuares a ler...
Methodology of the data preparation: 1. Data from each country survey was exported to IBM SPSS (using the syntax file provided by Limesurvey when exporting data to SPSS). 2. Qualitative variables Q00 and Q02 were automatically recorded into numerical form. 3. A binary variable “Completed” was created from the “submitdate”, using “recode into different variable” (Missing ->0, ELSE->1) to mark the complete answers from the incomplete 4. All string variables with Likert scales were transformed into numerical variables, using the “automatic recode” feature of IBM SPSS. Options ``Treat blank as user missing” and “Use same recording scheme” were activated. Then, all remaining codes for missing values or non-available data were coded with SYSMIS using “Recode into the same variable”. 5. The three questions S01c, L01c, and T02m had the scale of answers with an inverted logic. It was necessary to create new variables inverting the scale, using “Recode into different variables”. (1->5, 2->4, 4->2, 5->1, ELSE-> copy) 6. Measure of the numerical variables was corrected. 7. The labels of the questions were simplified (removing the answer options and non necessary text) 8. New variables for the constructs were computed using “Compute variable” with the mean of the variables that contribute to each construct in each group. (The syntax file “Data preparation.sps” can be used to automatize the procedure of data preparation. Just change the path in FILE=’…’ to your full directory path. The formulas for building the constructs are in lines 80-372.)
Info Adicional:
Methodology of the data preparation: 1. Data from each country survey was exported to IBM SPSS (using the syntax file provided by Limesurvey when exporting data to SPSS). 2. Qualitative variables Q00 and Q02 were automatically recorded into numerical form. 3. A binary variable “Completed” was created from the “submitdate”, using “recode into different variable” (Missing ->0, ELSE->1) to mark the complete answers from the incomplete 4. All string variables with Likert scales were transformed into numerical variables, using the “automatic recode” feature of IBM SPSS. Options ``Treat blank as user missing” and “Use same recording scheme” were activated. Then, all remaining codes for missing values or non-available data were coded with SYSMIS using “Recode into the same variable”. 5. The three questions S01c, L01c, and T02m had the scale of answers with an inverted logic. It was necessary to create new variables inverting the scale, using “Recode into different variables”. (1->5, 2->4, 4->2, 5->1, ELSE-> copy) 6. Measure of the numerical variables was corrected. 7. The labels of the questions were simplified (removing the answer options and non necessary text) 8. New variables for the constructs were computed using “Compute variable” with the mean of the variables that contribute to each construct in each group. (The syntax file “Data preparation.sps” can be used to automatize the procedure of data preparation. Just change the path in FILE=’…’ to your full directory path. The formulas for building the constructs are in lines 80-372.)
Autor:
Clica para continuares a ler...