The psychometric properties of the behavioral inhibition scale in a college-aged sample

Abstract

The Behavior Inhibition Scale (BIS) (Gest, 1997; Muris, Merckelbach, Wessel, & vand de Ven, 1999) has shown adequate psychometric properties in previous research using samples of children and adolescents. The purpose of this study was to examine the psychometric properties of the BIS in a college-aged sample of 112 females and 72 males. The results indicated that the BIS showed good internal consistency, was unitary and the scores were normally distributed. As predicted, BIS scores were significantly correlated with extraversion, neuroticism, self-reported depressive symptomatology and subjective well being. Contrary to previous findings, the BIS scores were not significantly correlated with self-reported anxiety symptomatology, and were more closely aligned with extraversion than neuroticism.

Introduction

Behavioral inhibition has been defined as a temperamental trait characterized by individual differences in shyness, fear and withdrawal reactions when encountering novel or unfamiliar situations (Kagan, 1997, Kagan et al., 1987). Previous research has shown that the behavioral inhibition construct, when measured in children and adolescents, relates to increased levels of negative emotionality, anxiety and depression (Biederman et al., 1993, Gest, 1997, Muris et al., 2003, Muris et al., 2001, Muris et al., 1999, Reznick et al., 1992, Turner et al., 1996). There is also evidence that behavioral inhibition, as a trait, appears to remain relatively stable from early childhood to adolescence (Gest, 1997, Kagan et al., 1988, Scarpa et al., 1995).

The Behavior Inhibition Scale (BIS) (Gest, 1997, Muris et al., 1999) is a short, self-report scale that was developed to measure the behavioral inhibition construct. Muris et al. (1999) administered the BIS along with the Screen for Child Anxiety Related Emotional Disorders (SCARED; Birmaher et al., 1997), the Penn State Worry Questionnaire for Children (PSWQ-C; Choprita, Tracey, Brown, Collica, & Barlow, 1997) and the Depression Questionnaire for Children (DQC; De Wit, 1987) to 152 children between the ages of 12 and 15. They reported correlations between the BIS and the other variables to be as follows: SCARED total score (0.46), PSWQ-C (0.41) and DQC (0.28). Among the SCARED subscales, significant correlations with the BIS ranged from 0.30 with the Generalized Anxiety Disorder and Separation Anxiety Disorder scales, to 0.78 with the Social Phobia scale. BIS scores were not significantly correlated with the following SCARED subscales: Obsessive-Compulsive Disorder, Traumatic Stress Disorder, Animal Phobia and Environmental-Situational Phobia.

Muris et al. (2001) conducted a similar study with 968 adolescents between the ages of 12 and 18. They reported significant correlations between BIS scores and the following variables: Children’s Depression Inventory (CDI; Kovacs, 1981) (r = 0.20) and the Spence Children’s Anxiety Scale total score (SCAS; Spence, 1998) (r = 0.36). All of the subscales of the SCAS were significantly correlated with BIS score with correlations ranging from 0.20 with the Obsessive-Compulsive Disorder scale, to 0.40 with the Social Phobia scale.

Muris et al. (2003) used a slightly modified version of the BIS with a sample of 327 adolescents between the ages of 11 and 15. They also obtained parent ratings of children on the BIS. Both child and parent reported BIS scores were significantly correlated with anxiety and depression as measured by an adapted version of the SCAS (Spence, 1998). However, parent reported BIS scores were not correlated with child reported depression symptoms. Also, the correlation between parent reported and child reported BIS scores was between 0.44 and 0.49.

Consistent with previous findings on behavioral inhibition, scores on the BIS have been shown to be related to negative emotionality, anxiety and depression (Gest, 1997, Muris et al., 1999, Muris et al., 2001, Muris et al., 2003), indicating that the BIS has demonstrated construct validity. In regards to scale reliability, Muris et al., 1999, Muris et al., 2001, Muris et al., 2003 reported the following BIS internal consistency coefficients: α = 0.82, 0.72 and 0.85. Thus, the BIS appears to be a psychometrically adequate measure of the behavioral inhibition construct in children and adolescents. However, there have been no studies reporting test–retest reliability of the BIS. Also, there have been no studies examining the relationship between the BIS and other established measures of behavioral inhibition.

Given the indications that behavioral inhibition appears to be a relatively consistent trait from early childhood to adolescence, it is logical to predict that the trait endures into early adulthood. The purpose of the current study is to examine the utility of measuring the behavioral inhibition construct with the BIS in a college-aged sample.

Eysenck’s (Eysenck & Eysenck, 1985) PEN model of personality indicates that there are three basic, orthogonal, personality dimensions: Psychoticism, Extraversion and Neuroticism. Extraversion reflects individual differences in social interest and activity level. Individuals who score high on Extraversion are sociable, lively, active, assertive, and sensation seekers, while low scorers are shy, quiet, passive, and careful. Neuroticism reflects differences in emotional reactivity to negative environmental stimuli. High scorers are prone to anxiety, depression, guilt, low self-esteem and moodiness, while low scorers are calm and less emotionally reactive. Gray (1987) indicates that behavioral inhibition, as defined in his theory, and clearly parallel to the construct under examination in this study, should be correlated with high levels of Eysenck’s Neuroticism factor, and low levels of Eysenck’s Extraversion factor.

• It is predicted that the BIS would demonstrate adequate psychometric properties in the sample of college students. Adequate psychometric properties will be operationally defined as an internal consistency coefficient over .70, a unitary factor structure and scores that are normally distributed.

• As indicated by Gray (1987) it is predicted that BIS scores would be significantly negatively correlated with Extraversion and significantly positively correlated with Neuroticism (Eysenck & Eysenck, 1985).

• It is predicted that BIS scores would be significantly negatively correlated with subjective well being.

• It is predicted that BIS scores would be significantly positively correlated with self-reported depressive symptomatology.

• It is predicted that BIS scores would be significantly positively correlated with self-reported anxiety symptomatology.