Survey research and design in psychology/Assessment/Lab report/Data recoding

Lab report - Reverse coding

After data screening, the following items need to be considered for reverse-coding.

Remember that it is a good idea to recode into different variables (rather than into the same variables).

Note that data recoding is also covered in exercises for each of the tutorials.

Perceived Stress Scale

edit

Items 4, 5, 7, and 8 are scored in the opposite direction to Items 1, 2, 3, 5, 9, and 10. Therefore, make sure to reverse-code one of these sets of items so that the variables are all measured in the same direction. Make sure that you understand what low or high scores indicate about stress. If the instructions below are followed.

Syntax:

* Reverse the code the negatively worded Stress item responses into new variables.
COMPUTE pss04r = 4-pss04.
COMPUTE pss05r = 4-pss05.
COMPUTE pss07r = 4-pss07.
COMPUTE pss08r = 4-pss08.
EXECUTE.

* Add variable labels for recoded variables.
VARIABLE LABELS pss04r 'felt confident about your ability to handle your personal problems (recoded)'.
VARIABLE LABELS pss05r 'felt that things were going your way (recoded)'.
VARIABLE LABELS pss07r 'been able to control irritations in your life (recoded)'.
VARIABLE LABELS pss08r 'felt that you were on top of things (recoded)'.

* Adjust decimal places for the recoded variables.
FORMATS pss04r pss05r pss07r pss08r (F1.0).

* Add value labels to the original variables. 
VALUE LABELS pss01 pss02 pss03 pss04 pss05 pss06 pss07 pss08 pss09 pss10
0 'Never'
1 'Almost never'
2 'Sometimes'
3 'Fairly often'
4 'Very often'.

* Add value labels to the recoded variables. 
VALUE LABELS pss04r pss05r pss07r pss08r
4 'Never'
3 'Almost never'
2 'Sometimes'
1 'Fairly often'
0 'Very often'.

* Compute composite score.
COMPUTE stress = mean (pss01,pss02,pss03,pss04r,pss05r,pss06,pss07r,pss08r,pss09,pss10).
EXECUTE.

Zimbardo Time Perspective Inventory

edit

No recoding needed, however the following value labels could be applied:

* Add value labels to the original variables. 
VALUE LABELS
ztpi01 to ztpi18
1 'Very untrue'
2 'Untrue'
3 'Neutral'
4 'True'
5 'Very true'.

Time Management Skills

edit

Responses to these variables needed recoding so that higher scores on all TMS items indicate stronger time management:

  • tm07 'I waste a lot of time.'
  • tm09 'I spend a lot of time mucking around.'
  • tm11 'I easily get distracted from important tasks.'
  • tm13 'I find myself procrastinating over tasks that need to be done.'
  • tm18 'I find distractions to be very tempting.'

Reverse code these variables using this syntax:

* Reverse the code the negatively worded Time Management item responses into new variables.
RECODE tm07 tm09 tm11 tm13 tm18
(8=1) (7=2) (6=3) (5=4) (4=5) (3=6) (2=7) (1=8) 
INTO tm07r tm09r tm11r tm13r tm18r.
EXECUTE.

* This is an alternative recoding method (achieves the same outcome - but will also deal with recoding any decimal values).
COMPUTE tm07r = 9-tm07.
COMPUTE tm09r = 9-tm09.
COMPUTE tm11r = 9-tm11.
COMPUTE tm13r = 9-tm13.
COMPUTE tm18r = 9-tm18.
EXECUTE.

* Add variable labels for the recoded variables.
VARIABLE LABELS
tm07r 'I waste a lot of time. (recoded)'
tm09r 'I spend a lot of time mucking around. (recoded)'
tm11r 'I easily get distracted from important tasks. (recoded)'
tm13r 'I find myself procrastinating over tasks that need to be done. (recoded)'
tm18r 'I find distractions to be very tempting. (recoded)'.

* Adjust number of decimal places for the recoded variables.
FORMATS tm07r tm09r tm11r tm13r tm18r (F1.0).
 
* Add value labels to the original variables. 
VALUE LABELS tm01 to tm18
1 'False - not like me'
2 'False - not like me'
3 'More false than true'
4 'More false than true'
5 'More true than false'
6 'More true than false'
7 'True - like me'
8 'True - like me'.

* Add value labels to the recoded variables. 
VALUE LABELS tm07r tm09r tm11r tm13r tm18r
8 'False - not like me'
7 'False - not like me'
6 'More false than true'
5 'More false than true'
4 'More true than false'
3 'More true than false'
2 'True - like me'
1 'True - like me'.