studies have shown that self-reports of attitudes and behaviour can be biased because of socially desirable responding (lajunen, corry, summala, & hartley, 1997; paulhus & reid, 1991). recent investigations have supported two distinct types of socially desirable response styles; impression management and self-deception. the present study evaluated the relationship between gambling behaviours and both forms of socially desirable response styles among social gamblers (n = 33), problem gamblers (n = 20), and non-gamblers {n = 22). three measures were administered: the south oaks gambling screen (lesieur & blume, 1987), the self-evaluation survey of gambling behaviour (beaudoin & cox, 1999) and the balanced inventory of desirable responding (paulhus, 1984). a small but significant negative correlation was found between impression management and problem gambling behaviours. significant negative correlations were found between impression management and numerous specific indices of problem gambling behaviour. contrary to predictions, no evidence was found to support the hypothesis that self-deception was a factor in problem gambling behaviour. socially desirable responses tended to be more frequent in the non-gamblers (impression management) and social gamblers (self-deception). additional analysis revealed sex differences in response patterns.