Individual differences in response scale use: Mixed Rasch modelling of responses to NEO-FFI items

Austin, Elizabeth; Deary, Ian J.; Egan, Vincent

doi:10.1016/j.paid.2005.10.018

Cited by 110 publications

(129 citation statements)

References 13 publications

Supporting

Mentioning

112

Contrasting

Unclassified

Order By: Relevance

“…In contrast, in the NERS class, threshold parameters were widely spaced, indicating that extreme categories only had the highest probability of being chosen at very high or very low trait levels. The same pattern was found in the English NEO-FFI (Costa & McCrae, 1992) by Austin et al (2006) and in several instruments assessing other constructs, such as a leadership performance scale (Eid & Rauber, 2000). According to Rost et al (1997, p. 331), ''The trait parameter of the mixed Rasch model is automatically corrected for the effects of a response set on the sum scores'' since person parameters are estimated separately for each latent class and class-specific item parameters (signaling the response style) are therefore taken into account.…”

Section: Methods Of Correcting For Extreme Response Style Effectssupporting

confidence: 62%

“…Data generation under the 1-dimensional and 2-dimensional models was conducted for the whole sample. Previous research indicates that some traits may be weakly to moderately associated with ERS (e.g., r = .22 with extraversion and conscientiousness; Austin et al, 2006). Thus, for the 2-dimensional data, two variations were simulated: one with a correlation of 0 between the trait and ERS and one with a correlation of .20 between the trait and ERS.…”

Section: Simulation Designmentioning

confidence: 99%

“…Response styles refer to systematic individual differences in response scale use that are characterized by a systematic preference or avoidance of certain response categories. If response styles exert an influence on item responses over and above the respondent's true latent trait level, trait estimates may be biased and in consequence inferences based on these trait estimates may be wrong (e.g., Austin, Deary, & Egan, 2006). This concern was already voiced rather dramatically by Cronbach (1950;p.…”

mentioning

confidence: 99%

“…Polytomous rating scales that are ubiquitous in the assessment of personality traits, attitudes, and many other constructs appear to be especially susceptible to response styles. Common response styles in self-report questionnaires applying polytomous rating scales include extreme response style (ERS), a preference for extreme response categories, and acquiescence response style, a preference for categories stating agreement (e.g., Austin et al, 2006;Baumgartner & Steenkamp, 2001;Bolt & Johnson, 2009). To quantify the influence of response styles on item responses, Wetzel and Carstensen (2015) computed differences in pseudo-R 2 values (Nagelkerke, 1991) between models ignoring response styles and two-dimensional item response models taking response styles into account.…”

mentioning

confidence: 99%

See 3 more Smart Citations

A Simulation Study on Methods of Correcting for the Effects of Extreme Response Style

Wetzel

Böhnke

Rose

2015

Educational and Psychological Measurement

View full text Add to dashboard Cite

The impact of response styles such as extreme response style (ERS) on trait estimation has long been a matter of concern to researchers and practitioners. This simulation study investigated three methods that have been proposed for the correction of trait estimates for ERS effects: (a) mixed Rasch models, (b) multidimensional item response models, and (c) regression residuals. The methods were compared with respect to their ability of recovering the true latent trait levels. Data were generated according to a unidimensional model with only one trait, a mixed Rasch model with two populations of ERS and non-ERS, and a two-dimensional model incorporating a trait and an ERS dimension. The data were analyzed using the same models as well as linear regression where the trait estimate is regressed on an ERS score and the resulting residual is considered the corrected trait estimate. Over all conditions, the two-dimensional model achieved the best trait recovery, though the difference to the unidimensional model was rather small. Mixed Rasch models were in general inferior to the other correction methods. When the trait and ERS showed no to weak correlations, ERS appeared to have a minor impact on trait estimation.

show abstract

Section: Methods Of Correcting For Extreme Response Style Effectssupporting

confidence: 62%

Section: Simulation Designmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A Simulation Study on Methods of Correcting for the Effects of Extreme Response Style

Wetzel

Böhnke

Rose

2015

Educational and Psychological Measurement

View full text Add to dashboard Cite

show abstract

“…These include issues such as how many points a scale should have (Krosnick, Judd, & Wittenbrink, 2005), whether it is advisable to use negatively (reverse) keyed items (e.g., Barnette, 2000;DiStefano & Motl, 2006), and whether various groups of people tend to respond to these items in different ways (e.g., Austin, Deary, & Egan, 2006). However, perhaps the most pressing issue is the fact that self-assessments are easier to fake than are other types of assessments (e.g., Griffith, Chmielowski, & Yoshita, 2007;Ziegler, MacCann, & Roberts, 2010).…”

Section: Self-assessmentmentioning

confidence: 99%

Teacher Leadership: An Assessment Framework for an Emerging Area of Professional Practice

Jackson¹,

Burrus²,

Bassett³

et al. 2010

ETS Research Report Series

View full text Add to dashboard Cite

This paper examines various definitions and frameworks that have been used to justify the emergence of a new category for the teacher professional: teacher leader. The emergence of this new professional category may lead to greater retention levels, and improved knowledge management and transfer within the teaching profession. Various key dimensions of this profession are examined, allowing us to highlight some key personal skills that would appear requisite for the teacher leader. An agenda for developing and validating assessments of teacher leadership is then proposed. It is argued that these assessments have the potential of legitimizing research within this field, as well as providing the opportunity to better understand what it takes to become a successful professional in this new domain of teaching practice.

show abstract

Stability of response styles in a personality measure: Evidence with clinical adolescents

2024

Couns and Psychother Res

View full text Add to dashboard Cite

BackgroundIn counselling and psychotherapy, individuals may exhibit response styles when completing personality measures, potentially distorting comparisons of inter‐ and intra‐differences. However, previous research has primarily focused on nonclinical samples and cross‐sectional data, leaving gaps in our understanding of the stability of response styles and their impact on clinical samples. This study aimed to fill gaps in knowledge by applying a mixed Rasch model to explore response style stability in a personality measure in clinical adolescents.MethodsThe study sample consisted of 290 clinical adolescents (112 males and 178 females) who were experiencing depressive symptoms and receiving continuous treatment for at least 2 years. Each participant completed a personality assessment at four time points with a six‐month interval between each assessment. To comprehensively examine the stability of response styles over an extended period, this study utilised mixed Rash modelling to analyse enduring response patterns using data collected at the initial (T1) and fourth (T4) assessments, which were spaced 2 years apart.ResultsUsing mixed Rasch modelling, the results showed that the extreme response style (ERS) and middle response style (MRS) were the two most common response styles in the five domains of the Big Five Inventory‐21 at T1 and T4. The ERS and MRS remained stable throughout the mental health intervention. There was no gender‐related relationship found for response styles in the data at separate time points (T1 and T4) or in the longitudinal data (long‐format data). Importantly, the findings revealed that odds of a clinical adolescent changing their personality response style (ERS → MRS or MRS → ERS) increased by 1.15 times with each unit increase in depression score.ConclusionResearchers and practitioners should be aware that the stability of the ERS and MRS for a personality measure may result in response style‐based explanations for findings obtained from longitudinal data, such as trait stability or trait change over waves.

show abstract

Individual differences in response scale use: Mixed Rasch modelling of responses to NEO-FFI items

Cited by 110 publications

References 13 publications

A Simulation Study on Methods of Correcting for the Effects of Extreme Response Style

A Simulation Study on Methods of Correcting for the Effects of Extreme Response Style

Teacher Leadership: An Assessment Framework for an Emerging Area of Professional Practice

Stability of response styles in a personality measure: Evidence with clinical adolescents

Contact Info

Product

Resources

About