Contrast of a 3-factor Structure for the Attitude Toward Lesbian and Gay Men (ATLG) scale
The aims of this paper were to compare the 3-factor model proposed by Moral and Valle (2011), for the ATLG scale, and provide validity evidences for it. The ATLG scale (Herek, 1984), the internalized homonegativity scale (HNI-16; Moral & Valle, 2013) and the homophobia scale (HFK- 10; Klamen...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Online |
| Language: | spa |
| Published: |
Consorcio de Universidades Mexicanas A.C.
2014
|
| Online Access: | https://psicumex.unison.mx/index.php/psicumex/article/view/221 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The aims of this paper were to compare the 3-factor model proposed by Moral and Valle (2011), for the ATLG scale, and provide validity evidences for it. The ATLG scale (Herek, 1984), the internalized homonegativity scale (HNI-16; Moral & Valle, 2013) and the homophobia scale (HFK- 10; Klamen, Grossman, & Kopacz, 1999) were applied to a non-probabilistic sample of 231 Mexican students of health sciences (121 women and 103 men). Three structural models were contrasted by Generalized Least Squares (GLS): one general factor, two correlated factors and three factors hierarchized to a general factor. The models of two correlated factors and three hierarchized factors had an equivalent goodness of fit and were significantly higher than the one-factor model. However, the data showed unidimensionality based on three arguments: 1) the values of correlations and regression coefficients of first-order factors were very high, 2) Horn’s parallel analysis indicated that the number of factors is one, and thus the variance explained by a second or third factor could be due to random, and 3) the goodness of fit of the unidimensional model was acceptable and very similar to that of the other two models. The correlations of the ATLG total score with the total scores of HNI-16 and HFK-10 were high. The self-defined sexual orientation and having gay friends had a medium size effect on the ATLG total score. It is concluded that the ATLG scale is unidimensional, and shows validity evidences. A simplified form of 5 items is suggested. |
|---|
