Oh G-d, A Borderline: Clinical Diagnostics As Fundamental Attribution Error Page: 27
The following text was automatically extracted from the image on this page using optical character recognition software:
these cases were re-coded as "eclectic". Two participants reported specific combinations of
ethnicities and were re-coded as "biracial." There were four participants who indicated
descriptions of their therapy hours (e.g., "multiple hundreds") that could not be easily transcribed
into a numeric value; these values were re-coded as missing and excluded from descriptive
statistical analyses of contact hours. For religion, the survey was set up in such a manner that
individuals who nominated "Christianity" were asked to provide a specific denomination; during
data preparation these nominal categories were all re-coded into the "Christianity" category.
Furthermore, two participants responded as "Other" and subsequently specified "Atheist;" these
individuals were re-coded into the "None" category.
Assumption of normality and outliers. Visual inspection of histograms of total scores on
the CAS - for the entire sample as well as by diagnosis and context conditions - did not indicate
outliers and suggested normal distribution. Furthermore the CAS total score variable did not
exhibit remarkable skewness or kurtosis- as measured by dividing skewness and kurtosis by
their standard errors- indicating adequate conformity to the assumption of normal distribution
(Tabachnik & Fidell, 1996). Following data preparation suggestions by Tabachnik and Fidell
(1996), z-scores were calculated for CAS total scores; no participant had a z-score greater than
3.3, indicating an absence of univariate outliers. The possibility of multivariate outliers was
investigated by calculating Mahalanobis distances for the CAS total score variable within the
four groups delineated by diagnosis and context. Z-scores of the Mahalanobis distances indicated
that no multivariate outliers were present (Tabachnik & Fidell, 1996).
Due to the true randomization of independent variables and the need to drop a few cases,
the group sizes for the ANOVA were not equal (Table 3). To evaluate for potential problems in
homogeneity of variance, an F-MAX test was conducted on the variances of the four groups
Here’s what’s next.
This thesis can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Thesis.
Schmalz, Jonathan. Oh G-d, A Borderline: Clinical Diagnostics As Fundamental Attribution Error, thesis, December 2011; Denton, Texas. (digital.library.unt.edu/ark:/67531/metadc103389/m1/33/: accessed August 18, 2017), University of North Texas Libraries, Digital Library, digital.library.unt.edu; .