Participation in athletics has benefits for persons with intellectual disabilities and their parents. Our purposes here were to confirm these benefits and to determine whether reports from athletes and parents were comparable (i.e., to test the validity of proxy responding). We conducted interviews with 34 Special Olympics sailing/kayaking athletes and their parents. Interviews took place at a regional Special Olympics competition and 12 months later by telephone. Findings suggest that Special Olympics participation has a positive impact on parents and athletes. However, relative to athletes, parents overstated the benefit, indicating that parental proxy responding should not be routinely used.
The value of sports participation for children and adults with intellectual and other developmental disabilities has been demonstrated empirically using qualitative (Farrell, Crocker, McDonough, & Sedgwick, 2004; Goodwin, Fitzpatrick, Thurmeier, & Hall, 2006) and quantitative (Dykens & Cohen, 1996; Gibbons & Bushakra, 1989; Wright & Cowden, 1986) techniques. Researchers have identified both physical and psychosocial benefits for athletes (Mactavish & Schleien, 2004; Weiss, Diamond, Demark, & Lovald, 2003) and for their parents and families (Goodwin et al., 2006; Klein, Gilman, & Zigler, 1993; Weiss, 2008; Weiss & Diamond, 2005). For example, for athletes, research has demonstrated improvements in various types of physical fitness and sports skills (Balic, Mateos, & Blasco, 2000; Castagno, 2001) as well as self-concept and self-esteem (Gibbons & Bushakra, 1989; Weiss et al., 2003). For parents with or without children who have intellectual and developmental disabilities, there is evidence of greater cohesion and adaptability in families who engage together in more leisure activity (Dodd, Zabriskie, Widmer, & Eggett, 2009). Several studies have also documented parental benefits, such as expanded social networks for parents and pride in children's accomplishments (Kersh & Siperstein, 2008; Weiss, 2008; Wiersma & Fifer, 2008).
The methods employed in the research cited have varied in many aspects, including who has been the primary informant providing the data. Although some researchers have focused exclusively on athlete participants (Dykens & Cohen, 1996; Gibbons & Bushakra, 1989), others have relied on parents to respond about their sons and daughters (Kersh & Siperstein, 2008; Klein et al., 1993; Megginson, Nakamura, & Furst, 1997). This latter technique of proxy responding raises the issue of response validity (e.g., Perry & Felce, 2002). Cummins (2002), in a thorough review of the research regarding proxy responding, concluded that some measures, especially those that require knowledge of another person's internal state (e.g., subjective well-being) should not be obtained by proxy.
Subsequent research has supported the Cummins (2002) conclusion. For example, Weiss and colleagues (2003) reported athlete, mother, and father ratings of general self-worth, physical competence, and social acceptance of adult athletes. For all three measures, mother and father ratings were similar to each other and lower than the athlete ratings in every instance, with five of the six comparisons statistically different. The remaining comparison was of borderline significance. However, Weiss et al.'s comparisons were not always from the same family groups. They reported data from 61 athletes, 71 mothers, and 42 fathers. Moreover, the response formats that these authors used may have been quite difficult for the athletes with intellectual and developmental disabilities (Finlay & Lyons, 2001). For example, the measurement of self-concept consisted of 21 items and a 2-step response sequence. First, the participants had to choose the competence statement that was “like” them and then indicate whether that statement was sort of true or really true. An indication of the difficulty was that 24% of the respondents did not give what were judged to be reliable answers. Although the authors believed that the remaining responses were reliable, it is possible that there were validity concerns about some of them.
We designed the current study with two primary purposes. First, we wanted to confirm the benefits of athletic participation for persons with intellectual and developmental disabilities as perceived by both athletes and their parents. Second, because of the concerns with regard to some of the methods used in previous research, particularly proxy responding, our goal was to test directly the comparability of parent and athlete perceptions about the value of sports participation for athletes. Specifically, we compared parent and athlete son or daughter pairs, so that parents were responding about their sons and daughters who were responding about themselves. Also, as described in detail in the Method section, we chose a response format that was consistent with recommendations for reducing bias (Finlay & Lyons, 2001) and that had already been used successfully with large numbers of individuals who had intellectual and developmental disabilities (Emerson & Hatton, 2008). Thus, our method should allow conclusions about the degree to which parental perceptions match those of their sons/daughters and, therefore, about the validity and usefulness of proxy responding.
Special Olympics athletes (n = 34) and parents of Special Olympics athletes (n = 29) from 26 different families participated in the original interview during a weekend sailing/kayaking regional event in Maryland. Table 1 presents the characteristics of the athletes and parents for the total sample as well as for a subset of 22 parent–athlete pairs. The paired subset is representative of the entire sample and not significantly different on any of the measured characteristics. The majority of the athletes were older than 22 years (range = 12 to 49 years), had mild disabilities, and were working at least part time. In addition to their participation in either sailing or kayaking, 65% of the subset sample participated in at least one other Special Olympics sport, with duration of involvement ranging from 1 to 25 years (M = 6.64).
Procedure and Materials
We collected data through 10- to 15-min semistructured interviews that were conducted in person and audiotaped with athlete and, if necessary, parent permission at the Special Olympics kayaking/sailing event. We adapted interview questions from previous studies of Special Olympics athletes and their parents (Kersh & Siperstein, 2008; Siperstein & Hardman, 2001), so that comparisons of our results with those of others would be possible. Interviews with parents focused on both their reactions and perceptions of their son's or daughter's sailing/kayaking participation. Although some questions provided response choices, the parent questions that are the focus of the current study—benefits/problems and expectations—had a free narrative response format. Parents also provided general family information and athlete-specific information, such as disability diagnosis and level, as well as schooling and employment status. Interviews with athletes focused on their own views of their Special Olympics sailing/kayaking participation, including why they had joined, what they did and did not like about the program, and their relationships with other athletes. Both parents and athletes were asked about athletes' happiness and feelings during sailing/kayaking participation and when not participating in sailing/kayaking. Visual cues for the measures of happiness and feelings were adapted from Emerson and Hatton (2008) and are presented in Figure 1. To indicate happiness, athlete interviewees were asked, “When you are (here now at/not at) sailing/kayaking. how do you mostly feel?” and then given the Faces card as a visual cue to choose one of the four options ranging from 1(broadly smiling with a verbal anchor of very happy) to 4 (broadly frowning with a verbal anchor of mostly unhappy). Questions for assessing feelings were of similar format: “When you are at/not at sailing/kayaking events how often do you feel _____?” and the visual cue card depicting feelings of sad/worried, left out, helpless, and sure of self was shown to the respondent. On a 3-point scale, possible answers ranged from a lot to almost never, with all items coded so that low scores indicated the presence of negative feelings or the absence of positive feelings, and high scores indicated the absence of negative feelings and the presence of positive feelings. The same procedures were followed with parents with appropriate modifications to the questions, so they were responding about their perceptions of their son's or daughter's feelings. The parent interview was also focused on parental perceptions of benefits to athletes and parental expectations of athletes.
In order to assess the stability of perceptions over time and obtain a behavioral measure of Special Olympics involvement, we recontacted 23 parents (88% retention) by telephone approximately one year later and asked whether their athletes would be participating in the sailing/kayaking event that year, and, if so, which family members would attend. In addition, parents were asked to indicate on a 4-point scale from none to a great deal and in narrative responses how Special Olympics in general (rather than sailing/kayaking specifically) had benefited the athlete, influenced the parent's expectations of the athlete, and benefited the family.
Data Analytic Strategies
We used the entire sample for all analyses, except the direct comparisons of parent and athlete responses when the paired subset was used. For this subset, our goal was to directly compare the responses of athletes and their parents to determine whether there were reliable differences between them, an outcome that would be relevant to assessing the benefits and risks of using proxy responding. Two independent judges coded all narrative responses, and we present percentage agreement in each case (see 3Results section). For different questions, the percentage ranged from a low of 64% to a high of 100%.
In general, parents reported that sailing/kayaking had a positive impact on family activities (64%), and no parents reported a negative impact. Nearly all parents (96%) reported that sailing/kayaking had a positive impact on their own social relationships outside the family, and 68% said that it had a lot of positive impact. No parents reported negative impact.
Benefits and expectations
All parents reported that their athlete had benefited from sailing/kayaking participation, and only 10% reported problems because of their athletes' sailing/kayaking participation. When problems were cited, the cause was generally not sailing/kayaking itself but, rather, some other aspect of participation (e.g., a failed romantic interaction with another athlete). Two raters coded narrative responses into five categories used by Kersh and Siperstein (2008) to denote athlete and parent perceptions of benefit to athletes: social, self-concept, participation, physical, and other. Table 2 displays the frequency and percentage of benefits in these five categories (intercoder agreement = 74%).
Further, 45% of parents in our sample reported that participation in sailing/kayaking influenced their expectations of their athlete (intercoder agreement = 76%); all of these parents indicated a positive direction of influence. Many of those who reported that sailing/kayaking had not affected their expectations for the athlete anecdotally indicated their reasoning for this response. These explanations give insight into contextual comparisons that parents made to assess their expectations. Some parents indicated that their expectations for their child in general contexts were “already high.” Others indicated that Special Olympics was a context in which they did not have expectations for their child's performance—participation was rewarding in itself and was over and above what they could ask for.
Happiness and feelings
Table 3 shows the results of the reports about happiness and feelings for the full sample of parents and for the paired subset of 22 parents and their sons or daughters. Both parent and athlete means reflected high levels of happiness and positive feelings in both contexts, but especially when they were participating in sailing/kayaking. Using the paired subset, we compared parent and athlete responses. For happiness, a Role (parent vs. athlete) × Context (in vs. not in sailing/kayaking) ANOVA, with context a within-subjects variable, revealed a significant main effect of context, such that both parents and athletes reported that athletes felt significantly happier while participating in sailing/kayaking, F(1, 42) = 27.59, p < .001, partial η2 = .40. Neither the main effect of role nor the Role × Context interaction was significant.
For the four feelings, we conducted a Role × Context × Feelings MANOVA, with repeated measures on context and feeling variables. Multivariate tests showed a significant main effect for Context, F(1, 38) = 14.63, p < .001, partial η2 = .28, and a significant interaction between context and role, F(1, 38) = 4.90, p < .05, partial η2 = .11, indicating that parents reported a larger difference for context than did athletes. Moreover, the interaction among context, role, and feeling approached significance, F(3, 36) = 2.62, .05 < p < .10, and there was a significant linear relationship for this interaction, F(1, 38) = 4.17, p < .05, partial η2 = .10. Univariate tests indicated that only the feeling of being left out had both a significant main effect of context, F(1, 40) = 8.00, p < .01, partial η2 = .17, and an interaction between context and role, F(1, 40) = 8.00, p < .01, partial η2 = .17. Specifically, parents reported that athletes felt significantly less left out during sailing/kayaking than athletes reported feeling during sailing/kayaking, t(42) = -2.31, p < .05.
Table 2 shows the benefits of Special Olympics to athletes as reported by parents. All parents in our sample reported that their athletes had benefited a great deal from their involvement in Special Olympics in general and noted at least one benefit. The majority of parents (87%) responded that they and their family had also benefited a great deal from their involvement in Special Olympics, whether as a spectator or at a more active level of involvement (e.g., volunteering with equipment setup, coaching, or records management). The narrative responses to the question about benefits to the athlete were coded into the same categories as those in the original interview (intercoder agreement = 64%) and are displayed in Table 2.
The narrative responses to the question about benefits to the family (intercoder agreement = 80%) were coded into a different set of five categories: family togetherness (different activities and increased amount of time spent together as a family), expanded networks (meeting other families, gaining social support or friendships from other families), values expression/enhancement (reflection of humanistic values in helping at Special Olympics, helping other families, feeling inspired by Special Olympics), new experiences (travel, new places), and positive-NOS (broad positive statements, not otherwise specified, for example, Special Olympics changing life, being a big part of life).
Family benefits most frequently cited by parents were values expression/enhancement (40%), followed by expanded networks (23%), family togetherness (19%), and the opportunity for new experiences (13%). Further, 6% of parents reported an overall positive benefit (positive-NOS) to their families from their involvement in Special Olympics.
Overall, coders had 100% agreement that parents' expectations for their athlete were fulfilled or exceeded: 74% of parents indicated that their athlete's participation in Special Olympics had influenced their expectations of their athlete a great deal (57%) or some (17%). In the instances where parents indicated their expectations had not been influenced (17%), they expressed similar reasoning compared to the responses in the original interview, such as participation alone was over and above anything they had hoped for.
At the original interview, all parents had reported that their athletes wanted to participate in the sailing/kayaking event again in the current year, and at the follow-up all but one, who had a conflicting activity, intended to do so. Most parents (83%) also planned to attend their athlete's upcoming sailing/kayaking event.
The results of the current study are consistent with those of previous research (Gibbons & Bushakra, 1989; Kersh & Siperstein, 2008; Weiss, 2008) in demonstrating the generally positive perceptions of Special Olympics participation both for athletes and for parents themselves. The large effect sizes for parents and athletes with regard to greater levels of happiness and more positive feelings and less negative feelings when participating in Special Olympics than when not indicates that these findings have practical as well as statistical significance. Moreover, these quantitative data were confirmed by narrative responses with respect to the specific event of sailing/kayaking at the time of the event as well as for Special Olympics more generally 12 months later. Behavioral confirmation of these narrative responses came with the large percentage of athletes who were planning to participate in the event again one year later.
Specific benefits that parents reported for themselves are consistent with those found in other investigations, including those with larger samples. For example, as with Kersh and Siperstein (2008), values expression/enhancement was one of the two most frequently mentioned benefits for the current sample both at the original interview as well as in the follow-up. Expanded social networks was also a benefit reported by parents in both studies. The support benefit of these social networks can be valuable, as has been demonstrated by Weiss (2008). Utilizing a sample of families whose children were Special Olympics athletes, he found that the mothers of children who had high levels of maladaptive behavior had less restriction of their roles if their children had higher levels of involvement in Special Olympics. Weiss explained this finding, in part, to be a result of the formal support that Special Olympics offers mothers.
In addition to confirming the findings of previous research, in the current study we report new findings in direct comparisons of parent and athlete responses to athlete feelings when they are or are not at a Special Olympics event. Although both athletes and parents were equally positive about happiness and positive feelings in the Special Olympics context, parents tended to overstate the value of Special Olympics relative to other contexts, especially for feeling left out. Parents reported the largest contextual difference for their athletes' feeling left out, whereas athletes reported no difference in context. It is possible that parents are responding to situational events that their sons and daughters did not notice or correctly construe and that their view is a more accurate depiction of reality. Nonetheless, given that these feelings are always subjective rather than objective, it is the athletes' reports that must be considered more valid.
This cautionary note with regard to using parental report as a proxy for athlete report echoes recommendations for best practices. Cummins (2002) reviewed more than a decade of research with regard to proxy responding for various measures of subjective well-being. His review included multiple types of samples, including those with intellectual and developmental disabilities. Cummins concluded that “subjective well-being cannot be validly measured by proxy” (p. 202). Holburn, Cea, Coull, and Goode (2007) made a similar recommendation specifically with regard to quality of life reports that differed for adults with intellectual and developmental disabilities and others responding for them.
On the other hand, individuals with severe levels of disability, especially when language and other communication is greatly impaired, may not be able to respond validly and reliably to subjective self-report measures. In a study of 154 randomly selected adults with intellectual and developmental disabilities in the United Kingdom, Perry and Felce (2002) found that two thirds of the respondents were either unable to respond to items or demonstrated marked response bias, suggesting that for participants with more severe impairments, it may become necessary to adopt alternative methods of data collection, such as the use of proxy responders. However, Perry and Felce also found a lack of concordance on subjective measures among those individuals who were deemed to provide valid responses and their proxies. This leads to an interesting dilemma; whereas the functional level of participants may at times necessitate the use of alternative methods, reliance on proxy report appears to be unsupported and may lead to spurious findings, particularly on subjective measures. Clearly, there is a need for greater exploration of alternative methods for measuring the subjective well-being and other similar constructs for individuals with intellectual and developmental disabilities.
Despite the value of these findings, some characteristics of the current study limit its usefulness. The sample was small, a characteristic feature of research in this field, especially for researchers who use an interview methodology. For example, although Mactavish and Schleien (2004) surveyed 65 families, they conducted interviews with only 16. Similarly, Goodwin et al. (2006), also using an interview methodology, included only 16 families in their sample. Nonetheless, even if typical, the current sample size and its homogeneity in terms of level of disability, race, and region of the country is limiting. More than two thirds of the athletes had mild disabilities, less than 10% of the parents were non-White or Hispanic, and all the families were from one state. Thus, we could not compare and contrast our findings across levels of disability or other athlete or family characteristics. Moreover, although many of the athletes participated in multiple sports, our original interviews focused on sailing/kayaking. We urge caution, therefore, in generalizing to samples that are more diverse than this one.
Furthermore, although our interrater agreement was quite good for most of the narrative response coding, an exception was the 12-month follow-up parent data for benefits of Special Olympics participation for athletes. This agreement was only 64%, a level considered marginal, in contrast to the more acceptable 74% agreement into the same five categories of benefits for the original interview questions (Cohen, 1960; Tinsley & Weiss, 1975). The 12-month follow-up, conducted by telephone, was briefer than the original interview, so fewer cues guided the scoring of responses. Moreover, an explanation for agreement not being higher at both measurements is that we used pre-existing categories rather than constructs that emerged from the narratives produced by the parents in our sample. Therefore, the obtained reliability may reflect the differences between our sample and the samples used by the investigators who originally developed the categories (Kersh & Siperstein, 2008). We used existing categories so that we would have the advantage of easy comparison of our results with those of other investigators. However, it is possible that by doing so, we introduced a somewhat lower reliability. Future investigators will need to balance the need for standardization with the value of good fit of sample and method.
Notwithstanding these limitations, this study is consistent with the results of other research. The benefits of athletic participation for both athletes with intellectual and developmental disabilities and their parents seem well-established for both segregated (Gibbons & Bushakra, 1989; Weiss et al., 2003) and integrated (Siperstein, Glick, & Parker, 2009) programs perceived at the time of the participation as well as over time and multiple participation events. The next generation of researchers should move beyond descriptive results and focus on exploring individual differences in the magnitude of the benefit and determine to what degree its effects spread to multiple facets of functioning.
This research was partially supported by faculty development grants from St. Mary's College of Maryland to the first author. We thank Special Olympics of Maryland for their collaboration in data collection and the families and athletes who participated.
Laraine Masters Glidden, PhD (email@example.com), Distinguished Professor of Psychology, St. Mary's College of Maryland. Department of Psychology, 18952 E. Fisher Rd., St. Mary's City, MD 20686. Katharine T. Bamberger, BA, Graduate Student, Department of Human Development and Family Studies, The Pennsylvania State University, College of Health and Human Development, 211 Henderson Building S., University Park, PA 16802-6504. Angela R. Draheim, BA, Departmental Assistant, Department of Psychology, St. Mary's College of Maryland, 18952 E. Fisher Rd., St. Mary's City, MD 20686. Joanne Kersh, PhD, Senior Research Associate, Center for Social Development and Education, University of Massachusetts, Boston, 100 Morrissey Blvd., Boston, MA 02125-3393.
Editor-in-Charge: Steven J. Taylor