Participation—defined as engagement in life situations, including leisure and recreational activities—is associated with the improvement of people with disabilities' quality of life. Several specific instruments assess leisure, but none of them has been adapted to the Spanish context. The goal of this study is to adapt and validate the Spanish version of the Leisure Assessment Inventory (LAI; B. A. Hawkins, P. Ardovino, N. B. Rogers, A. Foose, & N. Olsen, 2002). The adaptation of the original version of the LAI was carried out through translation and backward translation, and the validity of the instrument was analyzed. Descriptive analyses (means and standard deviations) were conducted for each LAI index. Construct validity was assessed through Pearson's product-moment correlation among the diverse LAI indexes, and convergent-discriminant validity through the correlation of the diverse indexes and the measures of quality of life. Results show that the LAI indexes are valid measures of the attributes of leisure behavior (participation, preference, interest, and barriers). This study provides a valid instrument to assess the participation profile of adults with disabilities in leisure activities.
Participation, an essential dimension of the International Classification of Functioning Disability and Health (ICF; World Health Organization, 2001), can be defined as engagement in life situations. Some examples of life situations in which adults usually participate include relations and interactions with friends and family; domestic life; learning and applying knowledge; and community, social, and civic life. Currently, the positive result of participation is considered one of the most important goals of the provision of services for people with intellectual disability (ID; Verdonschot, de Witte, Reichrath, Buntinx, & Curfs, 2009).
The model of the American Association of Intellectual Disability and Developmental Disabilities (AAIDD) in its 11th edition of Intellectual Disability: Definition, Classification, and Systems of Supports (Schalock et al., 2010) is consistent with the ICF model of human functioning. It defines ID in terms of human functioning, conceptualizing disability from an ecological and multidimensional perspective, and stressing the significant role played by individualized supports in the improvement of human functioning. The AAIDD manual considers participation one of the five dimensions of human functioning and defines it as “people's performance of life activities, and it is related to the person's functioning in society” (p. 10). Participation includes the social roles that are considered normal for a specific age group, for example, leisure activities. Research has shown that people with ID favor participation in leisure activities that include the community and that such participation improves their perception of quality of life and acquisition of adaptive skills (Cummins & Lau, 2003; Duvdevany & Arar, 2004; King et al., 2003; Kraemer, McIntyre, & Blacher, 2003; Orsmond, Krauss, & Seltzer, 2004). However, studies of participation also found that people with ID participate less in social and recreational activities compared to people without a disability (Braun, Yeargin-Allsopp, & Lollar, 2006; Duvdevany, 2002; Duvdevany & Arar, 2004; Orsmond et al., 2004; Poulsen, Ziviani, & Cuskelly, 2007) and that they often lack the necessary skills for social interaction and self-determination (Dattilo & Schleien, 1994; Duvdevany, 2008; Wehmeyer & Metzler, 1995).
Leisure can be understood as a distinct and major domain of life, similar to the domains of family, education, work, and participation in the community (Hawkins, 1994). AAIDD defines leisure as available free-choice time and those individually selected activities that characteristically are not related to work or other obligatory forms of activity that are expected to promote feelings of pleasure, affiliation, happiness, spontaneity, fantasy, or imagination, fulfillment, creativity, self-expression, and self-development (AAIDD, 2010).
Leisure behavior is operationalized as engagement in the leisure activity or repertory, expressed leisure preferences, interest in new leisure activities, and some measure of perceived freedom of choice (Hawkins, Ardovino, & Hsieh, 1998). The Leisure Assessment Inventory (LAI), developed by Hawkins and colleagues (1998), has four indexes based on this conceptualization of leisure. The Leisure Activity Participation Index (LAP) reflects the level of leisure repertory and is a measure of a person's engagement in the activity. The Leisure Preference Index (L-PREF) provides a measure of the leisure activities that please a person and that can be useful to increase participation. The Leisure Interest Index (L-INT) measures the person's level of interest in certain leisure activities in which the person does not currently participate or is impeded from participating in. And, lastly, the Leisure Constraints Index (L-CON) assesses the degree of internal and external barriers that impede participation in leisure activities. The validity and reliability analyses of the LAI were obtained through a longitudinal study of adults with ID (Hawkins et al., 1998). Construct validity of the LAI was determined through correlations among the indexes and the capacity of each index to discriminate between high and low performance. The results showed a low or null significance coefficient. Convergent and discriminant validity were confirmed by calculating the correlation coefficients with variables that may be related to leisure behavior. These variables included perceived life satisfaction and measures of adaptive behavior. Life satisfaction and adaptive behavior were measured with the Life Satisfaction Scale–Modified (LSS-M.; Hawkins, Kim, & Eklund, 1995) and the Inventory for Client and Agency Planning (ICAP; Bruininks, Hill, Weatherman, & Woodcock, 1986), the data of which were obtained in the third year of a longitudinal study (Hawkins et al., 1998).
The evidence of convergent validity was revealed through the significant correlation of the LAI with LSS-M. A significantly negative correlation was expected between LSS-M and L-INT, as well as between LSS-M and L-CON, because these indexes are sensitive to unsatisfied leisure needs. No significant relationships were found between LSS-M and L-PREF, because a preference for increasing participation in activities does not necessarily mean that a person is satisfied with life. Other findings that support convergent validity were obtained through significant positive correlations of the LAI indexes with the domains of adaptive behavior of the ICAP. The correlations among social and communication skills, life skills in the community, and the LAP index were significantly negative. The authors explain these results as the effect of acquiescence of people with ID with low social and communicative skills, which could affect their responses. Also, people with ID who had high life skills in the community were more self-determined in leisure activities, and their leisure repertory was more representative of their interests (Hawkins et al., 1998). Overall findings provide evidence of the instrument's stability, consistency, convergent and discriminant validity, and construct validity.
Currently, there is much literature that reveals the importance of people with intellectual and developmental disabilities' participation in leisure activities and its influence on their achieving a better quality of life; however, there is still no instrument adapted to and validated for the Spanish population. The LAI was developed to assess the leisure behavior of adults with IDD, and it can be used as a reference tool to plan leisure support services, which permits acquiring the skills involved in leisure, the promotion of physical and health activities, and the establishment and promotion of social networks, thereby facilitating social inclusion and promoting self-determination (Hawkins, 1993). This study focused on the translation, adaption and validation of the LAI into the Spanish language according to the International Test Commission Guidelines for translating and adapting educational and psychological tests (Hambleton, 1994).
A convenience sample of adults with IDD was obtained from diverse services that depend on the Federation of Organizations in favor of People with Intellectual Disability (FEAPS) and the Associations for Organizations in favor of People with Cerebral Palsy (ASPACE) in diverse regions of Spain. In total, 237 adults participated in this study, most of them (86.49%) had ID (70.46%); 16.03% had cerebral palsy with associated ID; 83.2% had disability rates higher than 65% (according to the system of examination, declaration, and classification of the degree of disability of Spain's Ministry of Health and Social Policy); more than one half (54.9%) had a moderate or severe level of ID; and only 14.8% took some kind of medication.
Of the participants in our study, there was a slightly higher percentage of men (52.7%) than of women (47.3%); their ages ranged between 17 and 64 years (M = 34.97, SD = 11.36); except for one interviewee, the rest of those who reported their civil status were single; when asked whether they had a partner, most of them said they did not (85.2%). With regard to the socioeconomic level, the majority were from a medium level (66%); they lived in urban settings (76.7%) and in the family home (56.1%). With regard to the educational level, a slight majority (54%) had attended regular school, and 70% had not exceeded primary studies. And, last, with regard to their work situation, one half had an occupation (50%). See Table 1.
The following instruments were used to collect the necessary data:
The psychologist of each center reported the age, place of residence, gender, civil status, educational level, residential modality, and type of intellectual disability of the young people and adults.
The Leisure Assessment Inventory (LAI; Hawkins, Ardovino, Rogers, Foose, & Olsen, 2002) is an instrument that was developed to measure leisure behavior in adults. The LAI has four indexes: (a) Participation in Leisure Activities (LAP), (b) Interest in Leisure Activities (L–INT), (c) Preference for Different Leisure Activities (L–PREF), and (d) Constraints to Leisure (L-CON). The inventory is applied in the format of an individually structured interview with a mean duration of 45 minutes, and it has three parts. The first part consists of questions about participation in 53 leisure activities. The second part has 20 closed questions about constraints to leisure participation, and the third part is made up of open questions about work, friends, family, and aging.
Two additional instruments were used to determine the validity of the four LAI indexes: the Escala integral de calidad de vida (Subjective Integral Quality of Life Scale; Verdugo, Arias, Gómez, & Schalock, 2009b) and the GENCAT Scale (Verdugo, Arias, Gómez, & Schalock, 2009a). Starting with the premise that the participation of people with ID in leisure activities is related to their perception of quality of life (Duvdevany & Arar, 2004), we used the scores in these instruments to explore the validity of the LAI.
The Subjective Integral Quality of Life Scale (Integral Subjective Scale) measures quality of life in people with ID both objectively and subjectively. This instrument is based on the theoretical model of quality of life proposed by Schalock and Verdugo (2002), which reflects the quality of life from the perspective of the person with disabilities and from the perspective of the professional.
The GENCAT Scale assesses objective quality of life in users of social services and is also based on the multidimensional model of quality of life proposed by Schalock and Verdugo (2002), and it provides a profile of the person's quality of life that serves to develop individualized support plans.
Design and Procedure
The design followed a cross-sectional study of the adaptation with translation and backward translation of the original version of the LAI. For the process of translating the instrument, the authors followed the recommendations for the adaptation of tests proposed by the International Test Commission (ITC; Beaton, Bombardier, Guillemin, & Ferraz, 2000; Hambleton, 1994). At all times, these recommendations for the process of cross-cultural adaptation of the measurement instruments were taken into account. Cross-cultural adaptation allows one to obtain instruments that are internationally comparable. In this sense, we followed a rigorous and systematic methodology to guarantee that the qualities of the adapted instrument would be similar to those of the original instrument.
The study was developed in five phases:
Three translators performed the translation.
A synthesis of the translations by consensus of the three translators, which produced the first Spanish version of the original instrument. Two of the items of the original version were modified because they did not match the Spanish context: Item 37: “playing shuffleboard” was changed to “playing petanque,” and Item 46: “playing softball or baseball” was changed to “playing soccer.”
The selection of two focal groups of 15 adults with ID, aged from 20 and 50 years, whose verbal and communication skills allowed them to understand and respond to the questions in order to analyze the cultural acceptability of the concrete questions, images, or phrases that had produced misgivings about their equivalence with the original version, and to obtain suggestions and ensure the suitability of the items of the instrument. For this purpose, we showed them the 53 photographs with the diverse leisure activities and asked them, “What activity is this person doing?” “Does this photograph correctly show the activity the person is doing?” “What would you change in the photograph?” The information collected by the participants of the two groups showed that each photograph correctly reflected the leisure activity.
Concordance and synthesis carried out by a committee of experts made up of four professionals from the area of disability to achieve semantic, idiomatic, experiential, and conceptual equivalence.
Backward translation carried out by two translators, whose maternal language was English and who were bilingual in Spanish, who met with the research team of the study, and together they compared the differences or discrepancies of both back-translated versions to the original version, obtaining as a result the definite version of the LAI.
The objective scale of the Integral Subjective Scale and the GENCAT Scale were completed by professionals who had known the participant for at least three months. The data of the LAI and the Integral Subjective Scale were collected in the form of individual interviews carried out by three field researchers, who received formal training from a research assistant before beginning the study. To avoid possible biases, none of the test applicators had any knowledge of the goals of the investigation, and each instrument was applied by different people to each participant. We obtained the informed consent of each participant and guaranteed the privacy of their data. This study received a favorable report from the Bioethical Committee of the University of Salamanca.
Statistical analyses were carried out with the SPSS program, version 18. Descriptive analyses (means and standard deviations) were conducted for each LAI index. Construct validity was assessed through Pearson's product-moment correlation among the diverse LAI indexes, and convergent-discriminant validity through the correlation of the diverse indexes and the measures of quality of life.
The scores on the LAP, L-PREF, and L-INT indexes were calculated with the 53 items about leisure activities presented in the scale. The score on the LAP index was the sum of the activities in which respondents stated that they participated. The L-PREF index was obtained by adding the preferred leisure activities from the list of activities they carried out. The total score of the L-INT index was the sum of the leisure activities in which the respondents did not take part, but which they would be interested in doing. Last, the L-CON index was calculated from 20 restrictions or barriers to participation in activities about which they had expressed interest but in which they had not taken part.
Table 2 shows the means, standard deviations, and ranges for each index. The mean number of leisure activities (LAP) carried out was 27 out of the 53 proposed, and 16 were preferred (L-PREF). Moreover, the participants expressed interest (L-INT) in participating in an average of 8 new activities, and they perceived an average of 6 barriers (L-CON) to their participation in them. According to the criteria of the original version of the instrument, a person's score is interpreted as high or low depending on whether it is one standard deviation above or below the mean for that index.
Validity of the Spanish Version of the Leisure Assessment Inventory Indexes
To assess construct validity, we used, on the one hand, the intercorrelations between the indexes and, on the other, the capacity of each index to differentiate between people with high and low levels of participation. For each index, the upper and lower 10% of the scores (as z scores) were compared with Student's t test for independent samples.
Specifically, as expected, we found a moderate and significant correlation (r = .549, p ≤ .01) between the LAP index and the L-PREF index because L-PREF is obtained directly from the activities in which the individual participates. However, this correlation was not high, which indicates that the L-PREF and the LAP index represent different dimensions of leisure behavior.
However, the L-INT index is not derived from the LAP index, and thus, we obtained a moderate, low, and negative relation between these two indexes (r = −.292, p ≤ .01). This confirms that the L-INT index measures a different dimension of leisure behavior. The remaining correlations were nonsignificant.
Student's t-test was calculated using the two subgroups of the sample (10% highest and lowest scores), and the results were significant: LAP index: t(47) = −31.586, p ≤ .01; L-PREF index: t(33.564) = −37.451, p ≤ .01; L-INT index: t(24.400) = −22.78, p ≤ .01; and L-CON index: t(46.878) = −36.009, p ≤ .01. These results provide evidence of the construct validity for each index to measure the different attributes of leisure behavior.
Correlations with Other Measures
To assess the convergent and discriminant validity of the indexes of the Spanish version of the Leisure Assessment Inventory, we examined the correlations of the LAI indexes with two scales that measure objective (GENCAT Scale) and subjective quality of life (Integral Subjective Scale; Verdugo et al., 2009b). Both instruments have good psychometric properties, based on the model proposed by Schalock and Verdugo (2002) and adapted and validated for the Spanish population.
We expected to obtain positive correlations between the LAP index and the different dimensions of quality of life: the higher the LAP levels, the higher the levels of quality of life. We found positive and significant correlations with the dimensions of the GENCAT Scale: Personal Development, Self-determination, and Social Inclusion (see Table 4).
However, we expected significant negative relations between the dimensions of quality of life and the L-INT and L-CON indexes because these two indexes indicate unmet leisure needs and problems to satisfy them, so higher scores in these indexes should be related to the perception of lower levels of quality of life. We obtained significant negative relations of the L-INT index with the dimensions of Interpersonal Relations, Material Well-being, Personal Development, and Self-determination of the GENCAT Scale (see Table 4). With regard to the L-CON index, we found significant negative correlations with the dimensions of Interpersonal Relations and Material Well-being of the GENCAT Scale (see Table 4) and with the dimensions of Physical and Emotional Well-being, Self-determination, and Material Well-being of the Integral Subjective Scale (see Table 5).
Last, we expected nonsignificant relations between the L-PREF index and the dimensions of quality of life, because a preference for certain leisure activities that the person is already carrying out should not be related to a worse perception of their level of quality of life. As expected, no significant positive relation was found with the dimensions of the Integral Subjective Scale (see Table 5).
These results support the relation between the diverse dimensions of leisure behavior, measured with the corresponding indexes, and the perception of quality of life. They also confirm that the diverse indexes assess different dimensions of leisure behavior.
This study analyzed the validity of the adaptation of the four indexes of the Spanish version of the LAI. Participation in family, school, work, and community is important to all people, and it has been shown to be related to a positive perception of quality of life. Although the concept participation in leisure activities is not new, the development of its conceptualization and its measurement still requires a further process of elaboration to clarify and operationalize it.
Regarding construct validity, these results show that the LAI indexes are valid measures of the attributes of leisure behavior (participation, preference, interest, and constraints). The LAP index presents an inverse correlation with the L-INT index, a similar result to that obtained in the original version of the instrument (Hawkins et al., 1998). The L-PREF index converged moderately with the LAP index, and the correlations among the other two indexes were nonsignificant, which shows that each index measures a different aspect of leisure.
Convergent and discriminant validity of the LAI indexes has been confirmed by the expected relations with measures of quality of life. Various studies have shown the influence of leisure activities in promoting quality of life (Duvdevany, 2008; Duvdevany & Arar, 2004; Iwasaki, 2007; McManus, Corcoran, & Perry, 2008). The LAP index was related to the objective perception of the different dimensions of quality of life (Personal Development, Self-determination, and Social Inclusion). In this sense, participation in leisure activities has been shown to predict the possibility of learning and personal fulfillment, deciding for oneself, being the causal agent of one's own life, and feeling like a member of society. Along these same lines, McGuire and McDonnell (2008) also found a positive correlation between the amount of time dedicated to leisure and self-determined behavior. In contrast, they found negative correlations between the L-INT index and the dimensions of Interpersonal Relations, Material Well-being, Personal Development, and Self-determination of the GENCAT Scale. These results suggest that functional skills (personal competence, adaptive behavior, and communication), economic resources, opportunities to learn, and the capacity for willful action do not predict the desire to increase participation in preferred activities. In the validation study of the original version of the LAI, the scores in this instrument were correlated with the LSS-M, and the same results were obtained (Hawkins et al., 1998), showing that the L-INT index measures a person's interest in certain leisure activities in which the person does not participate.
Likewise, an inverse correlation was found between the L-CON index and the objective quality-of-life domains of Interpersonal Relations and Material Well-being; that is, a larger number of relations with different people, having friends, and the resources to manage money lowered the tendency of the existence of environmental barriers to participation in leisure activities (Duvdevany & Arar, 2004). In the original study of the development and validation of the LAI, it was also found that the higher a person's life satisfaction, the fewer the environmental barriers to participation in leisure activities (Hawkins et al., 1998).
In this study, a significant correlation was found between the dimension Physical and Emotional Well-being of the subjective scale of the Integral Quality of Life Scale and the L-PREF index, indicating that life satisfaction, self-concept, and health increase participation in leisure activities that please the person. This result coincides with the findings of other investigations (Duvdevany, 2002; Duvdevany & Arar, 2004; Iwasaki, 2007). Likewise, this study has found inverse correlations between the subjective quality of life dimensions of Physical and Emotional of Well-being, Self-determination, and Material Well-being and the L-CON index, suggesting that feeling good and enjoying good health, the availability of money and work, and deciding for oneself may lead to decrease external and internal barriers that impede participation in leisure activities.
One limitation of the study was that it did not provide evidence of reliability of the four LAI indexes. In this sense, although we focused on the analysis of validity, reliability was analyzed to test measurement stability by applying the instrument at two different moments. However, the second measurement time was not very suitable because it corresponded to a time interval with different characteristics from the first one. Therefore, it would be necessary to retest the stability of the instrument in the near future.
The findings of the study indicate that the Spanish language version of the LAI is a valid instrument to measure leisure participation for adults with intellectual and developmental disabilities in Spain. Until now, no validated instrument adapted to our context was available; the fact of being able to adapt and specifically validate the instrument in the Spanish context turns the LAI into a specific instrument of our cultural context. However, this study does not provide information about the validity of the instrument in other Spanish-speaking cultural contexts; even so, it can be useful for other researchers interested in this emergent construct of study. Future research should consider the inclusion of Hispanic people living in the United States or in Latin American countries.
Ultimately, this study falls within the framework in the incipient efforts to develop robust measurement instruments, based on the social model of human functioning, and to provide information about participation of adults with IDD in leisure activities, which will serve to plan interventions and guide the processes of changing and improving the provision of such services. Therefore, the LAI contributes to the advancement of the study of participation in leisure activities from the viewpoint of youngsters and adults with IDD and from a multidimensional perspective of the construct leisure participation (leisure activity participation, preference, interest, and barriers).
Summing up, the results obtained in this study have shown that the relations between measures of quality of life and the LAI indexes support the validity of the instrument to assess the specific attributes of leisure behavior. This, together with its simplicity and feasibility, make it a useful instrument to assess the participation profile of the adults with disabilities in leisure activities, which will serve to plan for the person's support needs and to guide the professionals in the process of change and improvement of the leisure services.
Editor-in-Charge: Stephanie Dean
Marta Badia (e-mail: firstname.lastname@example.org), Institute of Integration on Community (INICO), Faculty of Psychology, University of Salamanca, Salamanca 37005, Spain; M. Begoña Orgaz-Baz, Miguel-Angel Verdugo, M. Magdalena Martínez-Aguirre, Egmar Longo-Araújo-de-Melo, and Ana M. Ullán-de-la-Fuente, University of Salamanca, Salamanca, Spain.