To show the results, please answer all the items and click „Apply a tape measure!“.
The results are based on a convenience sample of 4885 people (60% female) from the Czech Republic, aged 18–58 years (M = 23.3, SD = 4.3 years; 92% were between 18 to 28 years old). The data were collected through Facebook groups (more detailed description of the sample is provided in Rečka, 2018).
A bifactor model, in which all items loaded onto a general factor (explaining 48% of total variance) and only reverse scored items loaded onto a specific factor (explaining 11% of total variance) provied a good fit to the data. However, the model was not invariant across genders. Thus, men's and women's scores and not directly comparable. The fit of a configural CFA model, using MLR estimator, was: χ2(76) = 339.4, p < .001, TLI = .985, RMSEA = .038 (90% CI = [.034, .042]), SRMR = .019. Reliability estimates (McDonald's omega total) were ωt = .93 for men and ωt = .95 for women. Convergent validity was estimated as a correlation between the inventory sum score and self-reported height and equaled r = .86 for men and r = .89. for women. These values are used in the application.
Furthermore, construct validity was estimated using a structural model. The general and the specific factor together explained a substantial amount of variance in self-reported height of men, R2 = .81, and women, R2 = .84.
Sample description:
(shown after inventory completion)
This simple self-report inventory was created at the Deparment of Psychology,
Faculty of Social Studies, Masaryk University, Brno, Czech Republic. It serves as a tool
demonstrating the concepts of measurement in psychology and related fields and as a
teaching aid in the course „PSY259: Introduction to Psychometrics“.
Currently, the inventory results are based on Classical Test Theory. Therefore,
body height is predicted using a simple (raw) sum score of items, which can make the prediction
less accurate due to the non-linear relationship between the sum score and body height
(especially for very tall/short people) and due to the fact that the sum score is
less reliable than latent (factor) score estimates. However, both effects are negligible.
© 2019 Hynek Cígler, Deparment of Psychology FSS MU.
Additionally, Karel Rečka (2018) participated in item creation and collected the data that
are used for all the computation presented. Martin Tancoš (2019) carried out item selection
for the short version of the Brno Height Inventory
The source code of the application
is publicly available at github.