In my data, I have found that a measure of 'egoism' negatively correlates with social desirability (pearson correlation: -.19). Therefore, as an individual's social desirability increases, their self-reported egoism decreases (or vice-versa, as it is correlation).
I plan to use the egoism score in research relating to willingness to engage with pro-environmental behaviour, but I worry that without correcting for social desirability I may not uncover the 'true' influence of egoism.
Other than discuss this at the end of any article or paper, I wanted to know whether anyone had any experience, or had considered, pro-actively tackling the issue by 'correcting' people's scores for one variable based upon their score for social desirability? And if so, how they would suggest this could be completed?
As always, I would appreciate any thoughts or comments anyone has on the matter,
Thanks
Grant B
There is no simple answer to your question. SD can have several roles beyond simple bias, as measures of social desirability can also reflect personality. For example, Crowne and Marlow refer to SD as need for approval. If you statistically adjust for SD, you aren't necessarily removing a bias, but you might also remove other things as well. So results with SD controlled might be less accurate than results without controlling it.
That said, a correlation of -.19 is not very high, and likely it won't have much impact on your results. You can look at results both ways to see if it even matters. If results are different, you won't know why. I'll give you a few sources that discuss control in more detail in a general sense, and one having to do with negative affectivity as opposed to SD, but the issues are the same.
Atinc, G. M., Simmering, M. J., & Kroll, M. J. (2012). Control variable use and reporting in macro and micro management research. Organizational Research Methods, 15, 57-74.
Atinc, G. M., Simmering, M. J., & Kroll, M. J. (2012). Control variable use and reporting in macro and micro management research. Organizational Research Methods, 15, 57-74.
Becker, T. E. (2005). Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational Research Methods, 8, 274-289.
Becker, T. E., Atinc, G., Breaugh, J. A., Carlson, K. D., Edwards, J. R., & Spector, P. E. (2015). Statistical control in correlational studies: 10 essential recommendations for organizational researchers. Journal of Organizational Behavior.
Spector, P. E., & Brannick, M. T. (2011). Methodological urban legends: The misuse of statistical control variables. Organizational Research Methods, 14, 287-305.
Spector, P. E., Zapf, D., Chen, P. Y., & Frese, M. (2000). Why negative affectivity should not be controlled in job stress research: Don't throw out the baby with the bath water. Journal of Organizational Behavior, 21(1), 79-95. doi:http://dx.doi.org/10.1002/%28SICI%291099-1379%28200002%2921:1%3C79::AID-JOB964%3E3.0.CO;2-G
There is no simple answer to your question. SD can have several roles beyond simple bias, as measures of social desirability can also reflect personality. For example, Crowne and Marlow refer to SD as need for approval. If you statistically adjust for SD, you aren't necessarily removing a bias, but you might also remove other things as well. So results with SD controlled might be less accurate than results without controlling it.
That said, a correlation of -.19 is not very high, and likely it won't have much impact on your results. You can look at results both ways to see if it even matters. If results are different, you won't know why. I'll give you a few sources that discuss control in more detail in a general sense, and one having to do with negative affectivity as opposed to SD, but the issues are the same.
Atinc, G. M., Simmering, M. J., & Kroll, M. J. (2012). Control variable use and reporting in macro and micro management research. Organizational Research Methods, 15, 57-74.
Atinc, G. M., Simmering, M. J., & Kroll, M. J. (2012). Control variable use and reporting in macro and micro management research. Organizational Research Methods, 15, 57-74.
Becker, T. E. (2005). Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational Research Methods, 8, 274-289.
Becker, T. E., Atinc, G., Breaugh, J. A., Carlson, K. D., Edwards, J. R., & Spector, P. E. (2015). Statistical control in correlational studies: 10 essential recommendations for organizational researchers. Journal of Organizational Behavior.
Spector, P. E., & Brannick, M. T. (2011). Methodological urban legends: The misuse of statistical control variables. Organizational Research Methods, 14, 287-305.
Spector, P. E., Zapf, D., Chen, P. Y., & Frese, M. (2000). Why negative affectivity should not be controlled in job stress research: Don't throw out the baby with the bath water. Journal of Organizational Behavior, 21(1), 79-95. doi:http://dx.doi.org/10.1002/%28SICI%291099-1379%28200002%2921:1%3C79::AID-JOB964%3E3.0.CO;2-G
In general, I agree with Dr Spector; controlling for social desirability (sd) may remove variance which is not questionnaire response bias, but a general peronality trait. Also, the correlation you have found is probably not large enough to be very important.
Actually, I would rather like to point out that such a correlation may very well be due to common method variance (CMV) of other types than social desirability. Also, lie (sd) scales are not really well validated, so the results are not very trustworthy.
Today, it is very difficult to know whether sd and/or CMV will influence the results of a specific study. In my meta-analyses, I have reached very different concluions for different scales and variables. More research is needed on these issues.
Then again, if your measure of willingness to engage is not self-reported, but from some other source (I assume that egoism is self-reported), then you do not have a problem at all. The best solution is therefore to get data from a friend or relative on one of the variables involved, as that by-passes the whole problem.
Anyway, I would recommend these references;
af Wåhlberg, A. E., Dorn, L., & Kline, T. (2010). The effect of social desirability on self reported and recorded road traffic accidents. Transportation Research Part F, 13, 106-114.
af Wåhlberg, A. E., & Melin, L. (submitted). On the validity of lie scales.
af Wåhlberg, A. E. (2010). Social desirability effects in driver behavior inventories. Journal of Safety Research, 41, 99-106.
Uziel, L. (2010). Rethinking Social Desirability scales: From Impression Management to interpersonally oriented self-control. Perspectives on Psychological Science, 5, 243-262.
Podsakoff, P. M., Mackenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903.
Since a correlation of .19 explains less than 5% of the variance, I have to wonder why you are so concerned with it.
A bias does not need to be strong in an absolute sense to be important. The important thing is how strong it is in comparison to the associations you're interested in. In this specific instance which we are discussing, I do not think a .19 correlation is important, but this is mainly because this measurement is unreliable anyway.
- Badges
- Science topic
- Similar topics
- Psychology
- Social Psychology