Edmonds (1967) developed the Marital Conventionalization Scale to assess the contamination of marital adjustment scales by the social desirability response bias. Although marital conventionalization has been prominent in the marital quality literature, this study represents the first direct examination of Edmonds' hypothesis in a model testing framework. Confirmatory factor analytic results with 101 married participants did not support his conceptualization for males or females. An alternative model that treated marital conventionalization as another measure of marital satisfaction was supported in the confirmatory factor analyses. These results raise questions about the use of marital conventionalization scales as validity scales in marital quality measurement. Idealistic distortion is suggested as a more accurate term for this phenomenon. The vast majority of research on marital satisfaction relies on self-report measures. A persistent concern about the measurement of marital satisfaction has been the degree to which these assessments are contaminated by the social desirability response bias (Edmonds, 1967; Fowers, 1990; L'Abate & Bagarozzi, 1993). Edmonds (1967) developed the construct of marital conventionalization to refer to a tendency to endorse unrealistically positive descriptions of one's partner and marriage. He conceptualized marital conventionalization as a social desirability bias and argued that the distortion was unconscious and unintended. He developed the Marital Conventionalization Scale (MCS) with items that described the marriage in impossibly positive terms (e.g., My marriage is a perfect success, My spouse meets my every need). There has been considerable debate about the importance, place, and explanation of marital conventionalization. Subsequent research has confirmed that the MCS does correlate strongly (in the .50 to .73 range) with a variety of marital adjustment and satisfaction measures, partially confirming Edmonds' thesis (Edmonds, 1967; Filsinger & Wilson, 1984; Fowers, Applegate, Olson, & Pomerantz, 1994; Hansen, 1981). These findings have resulted in the recommendation to control for this bias in marital quality assessment (Fowers, 1990; L'Abate & Bagarozzi, 1993). Indeed, two of the most frequently used clinical inventories of marital satisfaction contain marital conventionalization scales (Olson, Fournier, & Druckman, 1987; Snyder, 1981). It has become increasingly clear, however, that the construct validity of the MCS is questionable. The MCS and marital adjustment measures have been found to have weak relationships to social desirability scales (Hansen, 1981; McCann & Biaggio, 1989). This is in contrast to the strong relationships among measures of marital adjustment, satisfaction, and conventionalization. In addition, Hansen (1981) presented correlational evidence that marital conventionalization may be another measure of marital adjustment. Two recent factor analytic studies provided stronger evidence for this, suggesting that marital conventionalization loads on the same factor as marital adjustment (Rychtarik, Tarnowski, & St. Lawrence, 1989) and marital satisfaction (Fowers et al., 1994). These studies have been inconclusive regarding Edmonds' (1967) hypothesis for three reasons. First, the factor analytic studies did not include social desirability measures, and therefore could not determine whether marital conventionalization represents a social desirability bias. Second, if what is usually termed marital conventionalization is another measure of marital adjustment or satisfaction, it would not be an ordinary scale. This is because the items reflect unreasonably positive perceptions of the marriage. Thus, if marital conventionalization is best considered another aspect of marital quality, it may be more accurately termed idealistic distortion. Third, Edmonds' (1967) conceptualization of marital conventionalization has not been adequately tested because he thought of conventionalization as self-deception. …