Garantía de los derechos sociales de los campesinos en Colombia

This chapter represents the first stage of meeting my aim to develop a self- report measure of RRBs for adults in order to assess the relationship between

imagination and RRBs in autistic adults. The specific aim of Study One was to test the RBQ-2A in NT adults. In addition to being a useful stage in the development and testing of RBQ-2A, Study One provided new data on the pattern of RRBs in NT adults. An adapted version of a parent report measure of RRBs, the RBQ-2A, was administered to a university student sample. PCA resulted in a two-component structure, one comprising motor behaviours, RMB, and the other behaviours related to routines and a preference for sameness, IS. As predicted, scores on the RBQ-2A were also

correlated with another measure of autistic traits, the AQ (Baron-Cohen et al., 2001); in addition all of the AQ subscales, with the exception of imagination, were

significantly associated with the RBQ-2A and its subscales. Finally, undergraduates scored significantly higher on RMB than IS.

The RMB component identified in Study One is similar to a component that is commonly found in previous research, RSMB (e.g., Cucarro et al., 2003; Leekam et al., 2007; Lidstone, Uljarević et al., 2014; Richler et al, 2010). Five of the six RMB items consistently load onto the factor that in previous research included motor and sensory items (RSMB), the exception being item one, arranging objects, which loaded here and in Leekam et al.’s (2007) analysis but not in Lidstone, Uljarević et al.’s (2014) analysis. Whereas it was predicted that the RBQ-2A would form two components, the results here were not entirely in line with predictions. The major difference between RMB

91

found here and RSMB in previous research is the lack of sensory items loading onto this component, which was not predicted. The second component corresponded to IS. This result was more comparable to previous research using the RBQ-2 in an ASD sample, with five items (13-17) loading in exactly the same way as in Lidstone, Uljarević et al.'s study.

In summary, the components yielded by the present PCA are similar to previous research with NT children and autistic children using the RBQ-2, with the exception of sensory items. Items two to six11 _{load onto RSMB in the child version of the}

questionnaire (Leekam et al., 2007; Lidstone, Uljarević et al., 2014) and RMB in the present study, and items 13 to 17 load onto IS across all three studies, supporting the construct validity of the questionnaire.

To my knowledge this the first study to look at the component structure of RRBs in NT adults, and so there are no analogous populations to compare them against. Therefore the most probable reason for the difference between the present PCA solution and previous research is that the present sample comprised NT adults whereas previous research examined NT children (Leekam et al., 2007) and children and adolescents with ASD (Lidstone, Uljarević et al., 2014). Certain types of behaviours may be associated with younger children or children with ASD rather than NT adults. For example, mean scores on items 3 (spinning) and 11 (carrying around objects) were higher in NT children (Arnott et al., 2010; Leekam et al., 2007) than in the present study. Moreover, autistic individuals show higher levels of sensory symptoms than NT individuals (e.g., Ben-Sasson et al., 2009; Kern et al, 2006; Leekam, Nieto, Libby, Wing, & Gould, 2007; Rogers & Ozonoff, 2005) and these items were not well endorsed by the present sample. Interestingly however, the participants scored themselves significantly higher in terms of RMB than in terms of IS. This may be because these items are tapping into behaviours associated with fidgeting and fiddling that are relatively common in typical populations.

11 _{These are: fiddling, spinning, rocking, pacing and repetitive hand/finger movements. See Appendix 3}

92

The different loading of certain items may also reflect the fact that certain behaviours do not clearly fall into one particular category. For example, eating a small range of foods (item 19) formed part of IS in the present study but has previously loaded on to RSMB (Lidstone, Uljarević et al., 2014) as well as IS (Leekam et al., 2007); eating a small range of food may be a result of sensory issues or insistence on

sameness and is therefore conceptually related to both subscales. The fact that these sensory items do not load in this analysis may reflect the slightly different wording of the questions. In particular, the sensory items 8 and 9 both ask about a special interest which is not a common or everyday term. As a result participants may be more

reluctant to endorse these items. However, the wording is the same for the parent version of the questionnaire in which sensory items do load onto a component (Lidstone, Uljarević et al.; Leekam et al.). It is not obvious why the wording of the sensory items would affect self-reporting NT adults differently to parents reporting on their children. Furthermore, in the next chapter, I present data from an online study of autistic adults in which the sensory items do load onto a component (page 112). Finally, of the eighteen items included in the analysis, the four items that did not load onto any component were answered on a three-point rather than a four-point scale. Given that the responses are collapsed before any analysis takes place, it is unlikely that this had any effect on the analysis; although it may be the case that the lack of a fourth response option resulted in more participants endorsing the mild/occasional response option as the less ‘extreme’ response. However, these same items are scored on a three-point scale in the parent-report version of the RBQ-2, and therefore this does not seem a likely explanation.

Although the primary goal of Study One was to assess the RBQ-2A’s reliability and validity in an undergraduate sample, the inclusion of the AQ allowed me to assess the relationship between imagination and RRBs, bearing in mind the criticisms I raised in the previous chapter regarding the imagination subscale of the AQ (page 77). Interestingly, in this sample the only subscale not associated with any of the RBQ-2A subscales was imagination. This supports Kloosterman et al.’s (2011) previous finding that imagination as measured by the AQ is associated with

93

undergraduate population. This does not necessarily impact negatively on the case for a relationship between RRBs and imagination; indeed Honey et al. (2007) identified a relationship between RRBs and play that was specific to autistic children.

There are some limitations in terms of the sample. Firstly, the sample

comprised only university students and is therefore limited in age and IQ distribution. However, this will be addressed by assessing more representative samples later in this chapter (page 94). A second limitation lies in the relatively small sample size for a PCA. There are no concrete rules for how large a sample size is needed for PCA (e.g.

Hogarty, Hines, Kromey, Ferron, & Mumford, 2005). Some argue that at least 100-300 cases are needed, whereas others recommend a specific ratio of cases to variables, ranging from 3:1 to 10:1 (e.g. Guadagnoli & Velicer, 1988; Tabachnick & Fidell, 2014; Williams et al, 2010). However, such rules of thumb have been found to be unreliable (e.g. Guadagnoli & Velicer, 1988; MacCallum, Widaman, Shaobo & Hong, 1999).

PCA was deemed appropriate here for the following reasons. First and foremost, the KMO measure of sampling adequacy was .79 which is well above the acceptability criterion of .5 (Field, 2013). The PCA also met all other assumptions of the analysis. Secondly, if communalities are above .5 then samples between 100 and 200 are adequate as long as there are few factors with a small number of salient variables (MacCallum, et al., 1999). Here there are two components with six and eight variables each, and all communalities were greater than .5 before rotation (M=.62). Finally, it is desirable for a component to have five or more loadings at .5 or greater (Osborne & Costello, 2009); however, if a component has four or more loadings greater than .6 then the pattern may be interpreted whatever the sample size

(Guadagnoli & Velicer, 1988). Here RMB has 5 item loadings above .6, suggesting it is reliable, although IS has just 3 loadings above .6 so does not meet this criterion. Nevertheless, both components meet Osborne and Costello’s (2009) criterion. Therefore, although the sample size is relatively small for PCA, given these strengths and the overall similarity of the PCA solution to previous research (with the exception of the sensory items), it presents only a minor limitation. Furthermore, the RBQ-2A was assessed in a larger sample, comprising participants with ASD, which will be reported in Chapter Four.

94

As part of collaboration with researchers in Melbourne, Australia, RBQ-2A data were available from adults with ASD diagnosed by a clinician and confirmed by the researchers, along with a group of NT adults matched on IQ. This enabled me to assess group differences with adults confirmed to have a diagnosis of ASD. In addition, the effects of IQ could be ruled out. Furthermore, when recruiting the NT control group, diagnoses of ASD were screened for as well as anxiety and mood disorders, allowing us more confidence in describing this group as NT. The next study assessed the

differences in RBQ-2A between these two groups as an additional measure of criterion validity.

3.3 Study Two: Comparison between ASD and NT participants