Based on discussion with Dr. Vésteinsdóttir, we expect the data to follow a normal distribution (Vésteinsdóttir, personal cimmunication, 2017). Because of this, we will run the analysis in an ANOVA.
We will first run a 2x2 ANOVA on the total score of social desireability. Each factor is speed (slow/fast) and answer orientation (true to the left/right).
This will be done using the following code:
BOOTSTRAP
/SAMPLING METHOD=SIMPLE
/VARIABLES TARGET=socdesire INPUT=fast trueleft
/CRITERIA CILEVEL=95 CITYPE=PERCENTILE NSAMPLES=10000
/MISSING USERMISSING=EXCLUDE.
UNIANOVA socdesire BY fast trueleft
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/EMMEANS=TABLES(OVERALL)
/EMMEANS=TABLES(fast) COMPARE ADJ(LSD)
/EMMEANS=TABLES(trueleft) COMPARE ADJ(LSD)
/EMMEANS=TABLES(fast*trueleft)
/PRINT=DESCRIPTIVE
/CRITERIA=ALPHA(.05)
/DESIGN=fast trueleft fast*trueleft.
We hypothesize that speeded judgments will lead to higher social desirability scores with no interaction with left/right orientation. However, there may be a main effect of left/right judgingness where speeded responses make people answer on the left side of the scale more often.
**References**
Vésteinsdóttir, V., Reips, U. D., Joinson, A., & Thorsdottir, F. (2017). An item level evaluation of the Marlowe-Crowne Social Desirability Scale using item response theory on Icelandic Internet panel data and cognitive interviews. *Personality and Individual Differences*, 107, 164-173.
