Not long ago, Adam Shaw (2012) blogged about two phenomena that might impact polling surveys, especially in the current presidential election in the United States. Named the "Bradley Effect" and the "Shy Tory Factor," as he describe them, they both seem to fall under a broad category known in social sciences research circles as "Social Desirability Response Bias." Coined by Crowne and Marlowe, back before 1960, SDRB reflects people's desire to "look good" when answering questionnaires. SDRB leads people to "over-report" what they think surveyors or other people around them desire to hear and to under-report what they think that others do not desire to hear. Although Mr. Shaw suggested that challenges have hindered or prevented pollsters from quantifying the existence of the the Bradley Effect and the Shy Tory Factor, other researchers have used several instruments that have allowed for the relatively easy confirmation of the presence of SDRB. One of these (the MC-Form C), derived from Crowne and Marlow (1960) by Reynolds (1982), includes only 13 "yes or no" items and has yielded about the same validity and reliability of the much longer, original instrument. Additionally, another of the "forms" proposed by Reynolds includes only 9 items, but it has presented lower validity. See more on Wikipedia: Social Desirability Bias.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24(4), 349-354.
Reynolds, W. M. (1982). Development of reliable and valid short forms of the Marlowe-Crowne social desirability scale. Journal of Clinical Psychology, 38(1), 119-125.
Shaw, A. (2012, August 30). How 'the shy republican' could be masking a landslide. [Web log post]. Retrieved from: http://www.americanthinker.com/2012/08/m-how_the_shy_republican_could_be_masking_a_landslide.html