Sample size and subject to item ratio in principal components analysis.

By Jason W. Osborne, Anna Costello. 2004

852 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/9a479236c3de1558473768dc2becdcc30fccadfa

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Psychometric Evidence of the Brief Resilience Scale (BRS) and Modeling Distinctiveness of Resilience from Depression and Stress (2018, background): “The number of DOI: 10.4236/psych.2018.97107 1847 Psychology cases per BRS indicator was multiple times the minimum requirements of 10:1 (Osborne & Costello, 2005), or 20: 1 (Schumacker & Lomax, 2015).”
Guidelines for developing, translating, and validating a questionnaire in perioperative and pain medicine (2017, background): “[52] Given the variation in the types of questionnaire being used, there are no absolute rules for the sample size needed to validate a questionnaire.[53] As larger samples are always better than smaller samples, it is recommended that investigators utilize as large a sample size as possible.”
The Relationship between Self-Confidence, Self-Efficacy, Grit, Usefulness, and Ease of Use of eLearning Platforms in Corporate Training during the COVID-19 Pandemic (2021, background): “The participantto-item ratio for this analysis was approximately 30 to 1, where the sample size was 307 and the number of variables included was 10; thus, the given sample size is sufficient to produce reliable results [67].”
Replication Analysis in Exploratory Factor Analysis: What it is and why it makes your analysis better (2012, methodology): “Indeed, this access to computing power has also allowed development of more complex estimation procedures (e.g., maximum likelihood) and the spreading use of confirmatory 1 Indeed, Henson & Roberts (2006) reported that approximately one-third of EFAs in high quality, measurement related journals…”
Using ΔGoodness-of-Fit Indexes in Assessing Mean Structure Invariance (2009, background): “=P ratio might be a more meaningful consideration than sample size alone (e.g., Gorsuch, 1983; Osborne & Costello, 2004).”
Using multi-item psychometric scales for research and practice in human resource management (2018, background): “The ideal minimum sample size for EFA has been debated frequently, with larger samples and larger item-to-participant ratios considered better (see Osborne & Costello, 2004, for a detailed discussion).”
The Impact of Culturally Engaging Campus Environments on Sense of Belonging (2016, background): “…of 1.0 are recommended for retaining principal components, while factor loadings of .40 and communalities of .50 are adequate thresholds for retaining survey items (see Hinton, McMurray & Brownlow, 2014; Larose & Larose, 2015; Osborne & Costello, 2004; Stevens, 1992; Tabachnick & Fidell, 2001).”
Energy at work: A measurement validation and linkage to unit effectiveness. (2012, methodology): “After we eliminated items with loadings of less than 0.40 and items that exhibited cross‐loadings greater than 0.32 (Osborne & Costello, 2004), 14 items remained.”
Why start a higher degree by research? An exploratory factor analysis of motivations to undertake doctoral studies (2015, background): “Although the ratio of subject-to-variable is an important factor to be considered before conducting an EFA, when total sample size increases, this ratio becomes less important (Osborne & Costello, 2004).”
A Validated and Reliable Instrument Investigating Engineering Students’ Perceptions of Competency in Generic Skills (2017, methodology): “…of the newly developed instrument, based on the ongoing debate on the appropriate sample size for factor analysis (e.g., Mundfrom, Shaw, & Ke, 2005; Osborne & Costello, 2004), it is arguable whether the sample size of 275 was reasonable for the validation of a questionnaire with 14 skill…”