Insufficient effort responding: examining an insidious confound in survey data

Area of Science:

• Psychological Measurement
• Survey Methodology
• Quantitative Psychology

Background:

• Insufficient effort responding (IER) is traditionally viewed as random error attenuating statistical associations.
• However, IER may introduce systematic bias, particularly affecting correlations between survey measures.

Purpose of the Study:

• To demonstrate how IER can systematically bias and inflate observed correlations.
• To propose that the mean score of attentive respondents (M_attentive) moderates IER's confounding effect.

Main Methods:

• Empirical studies using a personality questionnaire (Study 1) and an employee sample (Study 3).
• A simulation study (Study 2) to rigorously test the proposed mechanisms.
• Analysis of bivariate relationships between substantive measures in the presence of IER.

Main Results:

• Results consistently supported the hypothesis that IER can inflate correlations.
• The confounding effect of IER was negatively related to M_attentive, as predicted.
• IER was shown to inflate Type I error rates by strengthening observed relationships.

Conclusions:

• IER can systematically inflate correlations, challenging the assumption of mere attenuation.
• The impact of IER depends on the mean response level of attentive participants.
• Researchers and practitioners need to actively deter and detect IER to ensure data integrity.