Multilevel Factor Models for Ordinal Variables

Grilli, Leonardo; Rampichini, Carla

doi:10.1207/s15328007sem1401_1

Cited by 30 publications

(35 citation statements)

References 19 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In multilevel FA, separate factors are extracted for within- and between-person levels of measurement. The value of y * for item h of individual j on day i can be expressed as

y_{h i j}^{*} = μ_{h} + true[true \sum_{m = 1}^{M_{B}} λ_{m h}^{B} θ_{m j}^{B} + ε_{h j}^{B} true] + true[true \sum_{m = 1}^{M_{W}} λ_{m h}^{W} θ_{m i j}^{W} + ε_{h i j}^{W} true],

where M B indicates the number of between-person factors θ B with corresponding loadings λ B , M W indicates the number of within-person factors θ W with loadings λ W , μ is an item intercept, and the ε’s are item specific errors [for details, see 25]. …”

Section: Methodsmentioning

confidence: 99%

“…First, if the number of between- and within-person factors (M B and M w ) is not the same, then the items address different constructs across measurement levels, indicating a lack of dimensional invariance [26]. Second, if the loadings λ B and λ W differ, then the constructs have different interpretations on the between- and within-person levels, denoting “cross-level” DIF [17, 25]. …”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Psychometric characteristics of daily diaries for the Patient-Reported Outcomes Measurement Information System (PROMIS®): a preliminary investigation

et al. 2012

View full text Add to dashboard Cite

Purpose The Patient-Reported Outcomes (PRO) Measurement Information System (PROMIS®) has developed assessment tools for numerous PROs, most using a 7-day recall format. We examined whether modifying the recall period for use in daily diary research would affect the psychometric characteristics of several PROMIS measures. Methods Daily versions of short-forms for three PROMIS domains (pain interference, fatigue, depression) were administered to a general population sample (n = 100) for 28 days. Analyses used multilevel item-response theory (IRT) models. We examined differential item functioning (DIF) across recall periods by comparing the IRT parameters from the daily data with the PROMIS 7-day recall IRT parameters. Additionally, we examined whether the IRT parameters for day-to-day within-person changes are invariant to those for between-person (cross-sectional) differences in PROs. Results Dimensionality analyses of the daily data suggested a single dimension for each PRO domain, consistent with PROMIS instruments. One-third of the daily items showed uniform DIF when compared with PROMIS 7-day recall, but the impact of DIF on the scale level was minor. IRT parameters for within-person changes differed from between-person parameters for 3 depression items, which were more sensitive for measuring change than between-person differences, but not for pain interference and fatigue items. Notably, mean scores from daily diaries were significantly lower than the PROMIS 7-day recall norms. Conclusions The results provide initial evidence supporting the adaptation of PROMIS measures for daily diary research. However, scores from daily diaries cannot be directly interpreted on PROMIS norms established for 7-day recall.

show abstract

“…In multilevel FA, separate factors are extracted for within- and between-person levels of measurement. The value of y * for item h of individual j on day i can be expressed as

y_{h i j}^{*} = μ_{h} + true[true \sum_{m = 1}^{M_{B}} λ_{m h}^{B} θ_{m j}^{B} + ε_{h j}^{B} true] + true[true \sum_{m = 1}^{M_{W}} λ_{m h}^{W} θ_{m i j}^{W} + ε_{h i j}^{W} true],

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Psychometric characteristics of daily diaries for the Patient-Reported Outcomes Measurement Information System (PROMIS®): a preliminary investigation

et al. 2012

View full text Add to dashboard Cite

show abstract

“…Because our focus was only the latent structure of associate responses to FSSB items, a saturated latent covariance structure was specified at the supervisor level of the data structure. Given the complexity of MLCFA, we first conducted several suggested preliminary analyses (Grilli & Rampichini, 2007; Heck & Thomas, 2000) to assess whether a multilevel approach was needed and to identify measurement structure problems. For brevity, we only report the important findings from these preliminary analyses.…”

Section: Study 2: Fssb Superordinate Construct Validationmentioning

confidence: 99%

Development and Validation of a Multidimensional Measure of Family Supportive Supervisor Behaviors (FSSB)

Hammer

Kossek

Yragui

et al. 2008

Journal of Management

606

753

View full text Add to dashboard Cite

Due to growing work-family demands, supervisors need to effectively exhibit family supportive supervisor behaviors (FSSB). Drawing on social support theory and using data from two samples of lower wage workers, the authors develop and validate a measure of FSSB, defined as behaviors exhibited by supervisors that are supportive of families. FSSB is conceptualized as a multidimensional superordinate construct with four subordinate dimensions: emotional support, instrumental support, role modeling behaviors, and creative work-family management. Results from multilevel confirmatory factor analyses and multilevel regression analyses provide evidence of construct, criterion-related, and incremental validity. The authors found FSSB to be significantly related to work-family conflict, work-family positive spillover, job satisfaction, and turnover intentions over and above measures of general supervisor support.

show abstract

“…Technically, individual PANAS items should be treated as ordinal variables; however, because univariate skew and kurtosis were minimal, and identification and estimation of CFA models for multilevel data with ordinal variables is difficult and can lead to unstable factor solutions, this approach is permissible (see Grilli & Rampichini, 2007). …”

mentioning

confidence: 99%

Modeling trait and state variation using multilevel factor analysis with PANAS daily diary data

Merz

Roesch

2011

Journal of Research in Personality

142

125

View full text Add to dashboard Cite

This study used daily diary data to model trait and state Positive Affect (PA) and Negative Affect (NA) using the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988). Data were collected from 364 college students over five days. Intraclass correlation coefficients suggested approximately equal amounts of variability at the trait and state levels. Multilevel factor analysis revealed that the model specifying two correlated factors (PA, NA) and correlated uniqueness terms among redundant items provided the best fit. Trait and state PA and NA were generally associated with stress, anxiety, depression, and three types of self-esteem (performance, academic, social). The coefficients describing these relationships differed somewhat, suggesting that trait and state measurement may have different predictive utility.

show abstract

Multilevel Factor Models for Ordinal Variables

Cited by 30 publications

References 19 publications

Psychometric characteristics of daily diaries for the Patient-Reported Outcomes Measurement Information System (PROMIS®): a preliminary investigation

Psychometric characteristics of daily diaries for the Patient-Reported Outcomes Measurement Information System (PROMIS®): a preliminary investigation

Development and Validation of a Multidimensional Measure of Family Supportive Supervisor Behaviors (FSSB)

Modeling trait and state variation using multilevel factor analysis with PANAS daily diary data

Contact Info

Product

Resources

About