Methodological Issues

2.2.1 Methodological issues in past research

A lot of research work has been done in this field to look for the definition and measurement of self-enhancement. However, the operational definitions and measurement of self-enhancement has been proposed in diverse ways and result in different errors (Borkenau, Zaltauskas, & Leising, 2009; Colvin & Block, 1994a; Colvin & Block, 1994b; Colvin et al., 1995; Kwan et al., 2008;

Kwan, John, Kenny, Bond, & Robins 2004). Only by solving the measurement issue can we learn the precise results of self-enhancement.

First of all, self-report scale is one of the most commonly used methods for measuring self-enhancement, which measures the extent to which individuals report positive self-perceptions. However, it fail to distinguish “self-enhancers (people who perceived that they possess positive qualities when they do not)” from “accurate positive self-assessors (people who perceived that they possess positive qualities when they do) ” (Colvin & Block, 1994a; Colvin et al., 1995). For example, Rosenberg’s (1965) Self-Esteem Scale was one of the most frequently

10

used measures of self-enhancement. And, having high self-esteem in this scale does not necessarily means one has positive illusions. Thus, self-report scale did not have an accuracy criterion and it is important to measure self-assessment against the subjects’ actual ability, in order to analyze self-enhancement.

Second, social comparison and self-insight technique are also frequently used to measure self-enhancement and these do include criterions. However, Kwan and colleagues (2004; 2008) mentioned the limitations of the two methods. On the one hand, in the social comparison technique, self-enhancers were assessed as subjects who more positively assessed themselves than they assess their peers. This was affected by the perceiver effect: where an individual evaluates themselves more positively or negatively, than they assess other people. On the other hand, in the self-insight technique, self-enhancers were individuals who perceived themselves in a more positive way than other people perceived them. This did not remove the target effect:

where other people assess the individual more positively or negatively. These operational errors may result in divergent conclusions. Hence, “the inconsistent findings observed in the self-enhancement literature may, in part, reflect differences across studies in the amount of target and perceiver variance,” (Kwan et al., 2008, p. 1074).

Third, to remove the perceiver and target effects, Kwan et al. (2008) recommended the use of a componential approach based on the social relations model (SRM) in assessing self-enhancement. However, this approach requires each participant to rate every other participant in the sample. This is both costly and often impossible in studies with a large number of subjects.

Lastly, little is known about the underlying mechanism of self-enhancement and so it is measured in inconsistent ways. People may intentionally or unintentional present themselves in embellished ways. That is, self-enhancers are motivated to portray themselves as better (or worse) than is true. Kim, Zou, & Chiu (2010)’s pilot study tried to prove that participants possess some

11

knowledge of their performance, and their motivation to self-assess accurately on those tasks could be manipulated. Therefore, little research has been done on how to distinguish between individuals who self-enhance intentionally or unintentionally. The improved method will be elaborated in the following sections.

2.2.2 Methodological Remedies in the Current Research

The refined methodology in the present study was to address the motivational bias issue, by comparing the difference between participants’ perceived performance (with a monetary incentive) and actual performance. And, it reduced the errors that occurred from the previous researches.

First of all, different from self-report scales, this refined methodology includes an accuracy criterion which is the participants’ actual performance. Indeed, some researchers have proposed directly comparing participant’s self-appraisal of their academic performance with their actual performance to detect self-enhancing tendencies (Gramzow et al., 2008; Kwan et al., 2008;

Robins & Beers, 2001). And, in the present study, we compared the difference between participants’ perceived performance (with or without incentive) and actual performance. A positive difference score represents self-enhancement (unintentionally or intentionally), a negative score indicates self-effacement (unintentionally or intentionally), and a score of zero or close to zero indicated accurate or close to accurate self-assessment (unintentionally or intentionally).

Second, this refined methodology eliminated perceiver effects and target effects. On the one hand, by evaluating themselves on a specific measure (i.e., twenty math questions) rather than being evaluated by peers or a general trait (e.g., academic intelligence) the methodology can avoid perceiver effects. On the other hand, target effects can be avoided because individuals conduct self-assessments instead of being evaluated by others. Therefore, the method used in this

12

study was derived from the theories of SRM (Kwan et al., 2008), utilizes an accuracy criterion (i.e., actual performance), and removes concern over perceiver and target effects.

Lastly, by manipulating monetary incentive, this refined methodology addressed the motivational bias issue. Following Kim, Zou, & Chiu (2010)’s definition, we enriched their method with a monetary incentive. Assume that people are relatively accurate at judging their absolute performance level with a monetary incentive. The variation between perceived performance (with or with incentive) and actual performance reflect people's motivation to view themselves positively. The present study identified the motivation underlying the self-enhancing tendency and proved that some participants can self-assess accurately but choose not to.