Abstract
Objectives. The study assessed basic psychometric properties of Czech versions of the Big Five Inventory measuring five basic personality dimensions that differ with respect to number of items (BFI-44 and BFI-10).
Subject and settings. BFI-44 and BFI-10 for self-report were administered to six groups of 6.292 participants ageing from 11 to 84 years.
Hypothesis. Authors expected that Czech version of BFI would have good psychometric properties.
Statistical analysis. Internal consistency of BFI-44 scales was estimated using Cronbach alpha coefficient, internal consistency of BFI-10 was estimated using Spearman-Brown coefficient.
The test-retest stability for BFI-10 was estimated using Pearson's correlation coefficient. The structure of BFI-44 items was derived from Principal Component Factor Analysis with Varimax rotation.
The correspondence between BFI-10 and BFI-44 was estimated using Pearson's correlation coefficient. Results.
The BFI-44 scales showed adequate reliability, ranging from .65 to .83, Mdn = .73, across three samples of participants. The test-retest stability of BFI-44 estimated over a two month period was satisfactory, r = .79.
The test-retest stability of BFI-10 estimated over a month period averaged to r = .69, the test-retest stability estimated over a two month period in another sample averaged to r = .66. An average factor loading of BFI-44 items on intended factors was .50.
The BFI-10 scales captured 60% of the full BFI-44 variance after including an additional item for measuring agreeableness. Study limitation.
The study did not test convergent validity of instruments.