Data Analysis
The collected data were analyzed using the SPSS (Statistical Package for
the Social Sciences) for Windows 25.0 program. Descriptive statistics of
the data (frequency, percentage, mean, and standard deviation) were
calculated.
CVI was calculated for each item of the scale. CVI values were
calculated with the following formula: [(number of experts with rating
“completely suitable” + number of experts with rating “highly
suitable/minor revision required)/total number of experts]. Items with
CVI values greater than 0.80 were accepted to indicate sufficient
content validity.25
To test the construct validity of the scale, an exploratory factor
analysis (EFA) and a confirmatory factor analysis (CFA) were conducted.
The EFA and CFA were carried out on separate datasets. For the
separation of these datasets, the data points were listed as even and
odd numbers in SPSS. The points corresponding to odd numbers were used
for EFA (n=199), while those corresponding to even numbers were used for
CFA (n=199). The suitability of the data for factor analysis was tested
with the Kaiser-Meyer-Olkin (KMO) sample adequacy test and Bartlett’s
test of sphericity. A lower eigenvalue threshold of 1 and scree plots
were used to confirm the number of factors. Principal component analysis
was used as the factorization method, while varimax rotation, which is
an orthogonal rotation method, was selected as the rotation method. The
threshold of factor load values for each item was determined as a
minimum of 0.40.20,24,26 As model fit indicators in
CFA, χ2/df values smaller than five and RMSEA values
smaller than 0.08 are accepted to show a high goodness-of-fit. It was
accepted that CFI and GFI values needed to be greater than
0.09.20,27-30
The reliability of the scale was tested with Cronbach’s alpha and item
analyses. The response bias of the scale was tested using Hotelling’s
T². The homogeneity of the distribution of responses to each item was
tested to analyze response bias. This process involved checking whether
the mean scores of the items were significantly different from each
other. To determine whether the scale items were suitable for obtaining
a total score, Turkey’s test of additivity was carried
out.31-34