Patterns of Recreational Sport Participation within the Adult Population in Greece

Konstantinos Alexandris (University of Manchester Centre for PE and Leisure Studies)

 Abstract

While plenty of national and local surveys have been conducted world-wide with the aim of providing up to date data about sport and recreation participation, this issue has been widely neglected in Greece. The present study aimed, for the first time, to provide information regarding: (a) annual participation rates; (b) patterns of recreational sport participation; and (c) demographic differences (gender, age, education) in activities participation rates. The data were collected by means of a self-administered household questionnaire survey conducted in an urban area in Greece (city of Larissa). The results indicated that: (a) sport participation rates within the adult Greek population were considerably low; (b) walking and keep fit were the two most popular activities; and (c) different patterns of activities were revealed for the demographic groups. The implications of the results for the provision, planning and management of sport related services in the area in which the research was conducted are discussed.

Introduction

Physical recreation has been promoted in many countries through extensive provision of sport and recreation facilities and large publicity campaigns. This extensive promotion of physical recreation followed the recognition of multiple benefits of sport participation. Physical (Shephard, 1995; Sharkey, 1990), psychological (Berger, 1996) and social (Wankel & Berger, 1990) benefits have all been proved to be associated with physical recreation. These benefits are associated with reductions in medical and rehabilitative costs (Gratton & Taylor, 1985; Shephard, 1986) and improved performance in the workplace (Kimiecik & Lawson, 1996; Wallace, 1996).

Plenty of local and national studies have been conducted in many countries with the aim of providing data about patterns of sport and recreation participation. With reference to the situation in Greece, there is a lack of national and local data about sport participation. The aim of the present study was to provide research information, based on a local survey conducted in an urban area, about patterns of recreational sport participation within the adult population.

Background to the study

Studies in the field of sport and recreation participation can be classified in two categories: (a) those which made time-series analysis of data: These studies aimed to investigate changes in sport and recreation participation over time (e.g., Gratton & Tice, 1994; Robinson & Godbey, 1993); (b) those which made cross-sectional analysis of data: Annual participation rates and patterns of sport participation (e.g., Matheson, 1991), as well as demographic differences associated with sport participation (e.g., Coalter et al., 1995) were the main issues investigated by the cross-sectional studies. A few attempts have also been made to make direct comparisons between local participation figures across different countries (e.g., Stockdale, Wells & Rall, 1996; Gambetta & DePauw, 1995). The publication of Cushman, Veal and Zuzanek's (1996) book has been the first attempt to bring together cross-national data on leisure participation. Participation patterns of twelve countries were presented in this book, and according to Cushman, Veal and Zuzanek (1996) four main issues were addressed by the contributors: (a) the experience of national leisure surveys; (b) overall patterns of participation; (c) demographic differences in participation; (d) particular problems related to the conduct of national surveys.

Methodological Issues

The measurement of sport participation is one of the most important but also debated issues in the sport participation literature. Different measures have been used, all of which have both advantages and disadvantages. Four measures of sport participation have been used by the General Household Survey (Matheson, 1991), which is the major source of national data on sport and recreation participation in England. These are as follows: (a) Four weeks participation rates: the percentage of respondents who took part in an activity in the four weeks before the survey; (b) frequency of participation in a four week period: the average number of occasions of participation during the four weeks before the survey (c) frequency of participation per adult per year: the average number of occasions of participation during the last twelve months before the survey; (c) annual participation rates: the percentage of respondents who took part in an activity in the twelve months before the survey. The last measure (annual participation rate) was first introduced in 1987, in an effort to obtain measures of sport participation over a one year reference period, and to capture seasonal variations. Other measures of sport participation which appeared in the literature included:

(a) two weeks participation rates (Robinson & Godbey, 1993). The percentage of respondent who took part in an activity in the two weeks before the survey: (b) frequency of participation in a two week period (Robinson & Godbey, 1993). Number of times that the activity was undertaken in a two week period; (c) Annual frequency of sport participation measured ordinally. For example, "rarely", "sometimes", "often" and "very often" participation (Stockdale et al., 1996), and participation "more than four times", "one to four times" and " no-participation" (Kelly, 1980); (d) the average number of sport activities undertaken annually (Boothby, Tungatt, Townsend & Collins, 1981).

All these measures have both advantages and disadvantages. Measures of participation based on the one year reference period might not always be reliable and accurate, as respondents might have difficulties in recalling for such a long period (Veal and Cushman, 1996). Furthermore, as Chase and Harada (1984) suggested, self-reports of participation suffer from a substantial response error, that is the difference between actual and reported participation. Based on their empirical findings, Chase and Harada (1984) suggested that respondents usually exaggerate their frequency of participation. On the other hand, measures of participation based on the four week period fail to capture seasonal participation and infrequent participants. This is the main reason that annual participation rates are always higher than the four week rates (Matheson, 1991).

Aims of the Study

Considering the lack of data regarding recreational sport participation in Greece, and the important practical implications of research in this field, the present study aimed to investigate: (a) annual participation rates; (b) patterns of recreational sport participation; and (c) demographic differences (gender, age and education) in activities participation rates.

Method

The data were collected by means of a self-administered household questionnaire survey conducted in the city of Larissa. One thousand adult individuals were contacted of whom 50.2% completed the questionnaire (n=502). In order to contact the respondents ten streets were randomly selected and every fifth house in each street was visited, following Veal's (1992) suggestions. Primary and secondary students were excluded from the sample, as they participate in Physical Education programmes at their schools. Two main limitations of the sampling method should be addressed: (a) it was not a strict probability method (e.g., selection of respondents by electronic lists); (b) it was not very systematic. A stratified sampling method, for example, could have given a better probability of a more representative sample of the population. However, the method employed was the best choice considering the limited availability of time, and the limited financial resources.

Instrumentation

The instrument used was a self-administered questionnaire which comprised of two sections: 1) a sport participation checklist; 2) demographic questions. The sport participation checklist was based on a previous one developed specifically for the Greek population by Alexandris and Carroll (1997). Respondents were asked the extent to which they participated in recreational sport activities within the past year. Frequency of sport participation was measured ordinally. Responses were categorised along a continuum consisting of four response categories as follows: no-participation (non-participants), participation less than once a month (rare participants), participation at least once a month (moderate participants), participation at least once a week (frequent participants). Similar categories have been used by previous researchers (e.g. Howard & Crompton, 1984). As the aim was to measure the frequency of sport participation ordinally, the one year reference period was judged to be more appropriate than the four week period, which fails to account for seasonal participation and to capture infrequent participants (Matheson, 1991). However, the one year reference period also has limitations, which have already been discussed.

Respondents were also provided with a list of twenty two sporting activities and they were asked to indicate the extent to which they participated in any of these activities during the twelve months before the survey. These activities were reported by Alexandris and Carroll (1997) to cover the whole range of recreational sports undertaken by the Greeks in the area in which the investigation was conducted. Walking was included within the list. Since almost all activities involve walking, the term "walking for recreation and exercise purposes" was used.

In the second part of the instrument information regarding respondents' gender, age, and level of education was collected. The age was coded in ordinal categories, as follows: 18-25, 26-35, 36-45, 46-65. Four levels of education were included: primary level of education (individuals who were educated at the primary level), secondary level of education (individuals educated at the secondary level), university graduates, and university students. The demographic characteristics of the sample are presented in Table 1. A comparison between the sample obtained and the most recent local census which was available to the researcher (1991), indicated that the sample was almost balanced in terms of the gender groups. However, in terms of the age groups, the 46-65 group was shown to be under-represented in the sample, while the 18-25 and 26-35 age groups were over-represented. In terms of the level of education, respondents educated at the primary level were under-represented, while all the other education groups were over- represented. The biased sample might be related both to the limitations of the sampling method, as discussed previously, and the response rate (50.2%).

Results

Annual Sport Participation Rates

The results indicated that 68% of the sample were sport participants, while almost one third of the respondents (32%) never participated in any of the sporting activities during the twelve months before the survey. It should be pointed out that 23% of the participants stated walking as the only activity in which they participated in the twelve months before the survey. This suggests that if walking had not been included within the list of the sporting activities, the non-participants would have been the majority in the sample (55%). As discussed previously, the sample of the study was not shown to be fully representative of the adult population in the area in which the investigation was conducted. Non-educated and older individuals were shown to be under-represented in the sample. In order to indicate an example of the effects of this biased sample on the results produced, the annual participation rate based on a weighted sample was also calculated. The sample was weighted according to the 1991 local census, which was the only one available when the research was conducted. The analysis indicated that the annual participation rate fell from 68% (unweighted sample) to 43.5% (weighted sample). The second figure (43.5%) seems to be closer to the actual participation rate of the study population (population of the city of Larissa).

Frequency of Sport Participation

In order to examine the frequency of sport participation, the proportions of participants who fell into the three categories (frequent, moderate and rare participants) were calculated. The results indicated that there was a certain group of individuals (27%) who were shown to be committed to an active lifestyle (participation on a weekly basis, frequent participants). The rare participants (participation less than once a month) were found to be the second biggest participant group, representing the 24% of the sample. Finally, 17% of the respondents stated that they participated in sports on a monthly basis (at least once a month, moderate participants). Both the moderate and especially the rare participants can be considered as infrequent participants, and in fact they were shown to be the majority in the sample. Together, they counted for 41% of the whole sampling population.

Annual Activities Participation Rates

In terms of the individual activities participation rates, walking was shown to be the most popular activity (Table 2). More than 30% of the respondents stated that they went walking for exercise purposes at least once in the last twelve months before the survey. Keep fit was reported as the second most popular activity, with a participation rate of 29%, followed by jogging (participation rate 22%). Three activities, basketball (21.5%), weight-training (21%) and aerobics (20.7%) were undertaken by 20% or more of the respondents. The considerably high participation rate of basketball (21.5%), which was reported as the most popular team activity, is worth pointing out. The participation rates of all the individual sporting activities included in the list are presented in Table 2.

Patterns of Sport Participation

In order to identify the underlying structure of sport patterns based upon intercorrelations among participation rates for the relevant activities, a principal component analysis was performed using the SPSSx Factor (SPSS Inc., Chicago, USA) programme. According to Stockdale et al. (1996), the existence of low frequency activity categories creates a problem for the application of a factor analysis technique. For this reason a 5% participation rate was proposed as a criterion for including an activity in the analysis. Following this criterion fourteen activities were included in the principal component analysis.

Only those components with eigenvalue greater than 1.0 were retained and rotated with both orthogonal and oblique rotation. Orthogonal rotation was retained because no significant correlations between the factors were revealed, when oblique rotation was performed. Four factors emerged which accounted for 51.7% of the variance, as follows: (1) factor 1 was the most clear of the four, comprised of the four typical fitness activities, weight-training, keep fit, aerobics and jogging; (2) factor 2 comprised of three outdoor activities, walking, hiking and cycling, together with swimming; (3) factor 3 comprised of two team activities, basketball and football, together with athletics; (4) factor 4 comprised of tennis, volleyball and dancing. It is worth noting that four activities loaded on more than one factor. Jogging also loaded (.37) on factor 3 (football / basketball), cycling also loaded (.41) on factor 4 (tennis / volleyball), athletics also loaded (.43) on factor 1 (weight-training / keep fit), and dancing also loaded (.43) on factor 4 (tennis / volleyball).

Activities Participation Rates by Gender

Results regarding gender differences in activities participation rates are presented in Table 4. In order to make the analysis meaningful (adequate number of respondents within the cells) only activities with participation rates of more than 10% were included. Statistical significant differences were found in five of them, aerobics (x2=46, p<.001), basketball (x2=46, p<.001), keep fit (x2=6.7, p<.01), jogging (x2=4.1, p<.05), and dancing (x2=5, p<.05). Aerobics, keep fit and dancing were the two female dominated activities, while basketball and jogging were dominated by males. The "top ten" most popular activities for males and females are presented in Table 3. Basketball (33%) and football (32%) were the two most popular activities for the males. In contrast, two individual activities, keep fit (34%) and aerobics (32%), were the two most popular activities for the females. Walking was shown to be the third most popular activity for both gender groups.

Activities Participation Rates by Age

Results regarding age differences in activities participation rates are presented in Table 4. Statistical significant differences were found in eight of them, volleyball (x2=56. p<.001), aerobics (x2=24, p<.001), basketball (x2=21, p<.001), dancing (x2=21, p<.001), weight-training (x2=15, p<.005), keep fit (x2=13, p<.005), jogging (x2=13, p<.005) and swimming (x2=9, p<.05). As shown in Table 4, there was a clear decline in participation rates of all these activities (with the exception of dancing) with advancing age. This decline is dramatical in volleyball and aerobics. The "top ten" most popular activities for the four age groups are presented in Table 4. Walking was the most popular activity for three of the four age groups (26-35, 36-45 and 46-65). The popularity of walking was shown to significantly increase with advancing age. It is worth noting that walking was by far the most popular activity for the age group 46-65 (36%). The participation rate of the second most popular activity in this group was much lower (keep fit =17%). Keep fit was the most popular activity for the 18-25 age group, with a participation rate of 39%, and the second most popular activity for all the other age groups. Finally, the youngest age groups (18-25 and 26-35) seemed to be more attracted than the other two groups to physically demanding and competitive activities, such as jogging, aerobics, volleyball and basketball.

Activities Participation Rates by Level of Education

Results regarding age differences in activities participation rates are presented in Table 5. Statistical significant differences were found in seven of them, volleyball (x2=49, p<.001), basketball (x2=22, p<.001), walking (x2=14, p<.005), jogging (x2=14, p<.005), keep fit (x2=14, p<.005), aerobics (x2=12, p<.01) and weight-training (x2=9, p<.05). In all these activities there was a trend towards decreased participation rates within the group with the lowest level of education (primary group). Students' significantly higher participation rates in volleyball (49%) and basketball (41%) are worth noting. The "top ten" most popular activities for the four educational groups are presented in Table 5. Walking was shown to be the most popular activity for the primary and the university graduate groups (17% and 42% respectively), while keep fit was the most popular activity for the two remaining groups (university students 43% and secondary group 27%). It should be pointed out that half (50%) of the participants in the primary group reported walking as the only activity that they participated in during the twelve months before the survey. The low participation rate of swimming among the individuals within the primary level (4.2%) and the university students (13%) are worth noting.

Discussion

The results of the present study provided evidence that participation rates in recreational sport activities within the adult population in Greece are low. Almost one third of the respondents (32%) were shown to be absolutely inactive in terms of sport participation. As discussed previously, walking "for exercise purposes" was included within the list of the recreational sports, although it is not a clear sporting activity. It is worth emphasising that the results indicated a dramatic decline in the annual participation rate when walking was excluded. A significant proportion of the population (23%) reported walking as the only activity in which they participated in the year before the survey, which suggests that if walking had been excluded from the list of the sporting activities, the proportion of the sport participants would have been fallen to 45%. Furthermore, as discussed previously, the sample of the study was not shown to be fully representative of the adult population in the area in which the investigation was conducted. This is an important issue which should be considered in any attempt to make generalisations. Annual participation rates based on a representative sample of the population might have been lower. In order to support this argument an example of the annual participation rate, based on a weighted sample (according to 1991 census), was presented. The annual participation rate fell from 68% (unweighted sample) to 43.5% (weighted sample). The second figure (43.5%) seems to be closer to the actual participation rate of the study population (population of the city of Larissa). This figure is very low and suggests that the majority of the adult population in the area (56.5%) abstain from any kind of physical recreation. Unfortunately, in the absence of any national or local data related to sport participation in Greece, it is impossible to compare these findings and to draw more definite conclusions. However, this proportion of non-participants is much higher than that those found from national surveys in other countries (e.g., England, 24.5%, Matheson, 1991; France, 25%, Samuel, 1996). Finally, it should be pointed out that the study was conducted in an urban area. It is estimated that participation rates in rural areas would be even lower (although further research is required in order to support this argument).

The participation rates of basketball and swimming are worth discussing. Basketball (participation rate 21.5%) was shown to be the most popular team activity and the sixth most popular activity of all included in the list. In contrast, swimming (14.2%) was not included within the "top-ten" most popular activities, which was, to a degree, unexpected, considering the climatic conditions in Greece and the high participation rate of swimming in other countries (e.g., 34.6% in England). It could be argued that these two activities participation rates might be closely related to the way that sports are promoted by the central government. They are indicative examples of how individual sports can be effectively promoted (basketball) or widely neglected (swimming) by the central government. The Greek Basketball Association (GBA) is one of the few governing bodies of sport, which received increasing funds by the General Secretariat of Sport over the period 1988-1991 (Nassis, 1996). With the aim of promoting basketball and increasing its popularity among the public, the GBA implemented a number of strategies, such as extensive promotion through the media, close co-operation with local authorities and schools, and provision of public outdoor basketball courts in the last ten years. Most of the local authorities (among them the Sports Development Department of the municipality of Larissa) established the organisation of local basketball leagues for the public in an effort to introduce the sport to the public and increase its popularity. Further to this, all schools (primary and secondary) were provided with outdoor basketball facilities, and local and national basketball leagues at school level were organised. All these, together with the national team's success in international competition contributed towards a rapid increase in the popularity of basketball. In contrast, recreational swimming has been neglected by the central government, although it has been promoted world-wide (e.g., England; Sports Council, 1994; Taylor, 1995) as an ideal recreational sporting activity. It is worth examining briefly the opportunities provided to the public for participating in swimming and water activities in the area, as these might relate to the low participation rate of swimming. In terms of the provision of facilities, there is only one public swimming pool in the area of the investigation (city with 150,000 population). Furthermore, there were no supervised sessions available for adults. According to the Sport Development Department of the city of Larissa (1995), the swimming pool was not available to the public on a long-term basis, and it was to be used only by the local swimming clubs. These clubs, provide teaching sessions only for young swimmers in an effort to promote sports for excellence. Consequently, the private leisure pools were for a long time the only choice available to the public. Furthermore, it is worth noting that swimming is not among the sports taught to pupils in the primary and secondary schools. All these factors suggest that the low participation rate in swimming may not be due to lack of interest or demand for swimming, but might be related to the limited opportunities for participation.

In terms of the patterns of the individual activities, the results of the factor analysis indicated that the sporting activities do form groups, which are relatively independent of one another. Individuals who participate in weight-training sessions were shown to be likely to participate in keep-fit, aerobics sessions and to go jogging as well. Keep fit, aerobics and weight-training are the three typical fitness related activities, which in Greece are mainly provided by private health clubs. Three outdoor activities, walking, hiking and cycling, together with swimming formed the second group of activities. The main characteristics of the three first activities are that they are non-competitive and outdoor activities. The two most popular team activities, basketball and football, together with athletics formed the third category of activities. These activities are competitive and are mainly provided by the public sector (outdoor sport facilities). Finally, individuals who played tennis were also shown to be likely to play volleyball and to participate in dancing sessions (fourth group of activities). It is difficult to compare the factor structure identified in the present study with those found in other countries. Stockdale et al. (1996), in one of the few studies in which a factor analytical design was employed in an effort to categorise activities, reported eleven groups of activities, based on English data, and twelve groups of activities, based on American data. However, this study included a wide range of leisure activities (free time activities), and the findings cannot be compared with the findings of the present study.

In terms of the frequency of sport participation, it is quite encouraging that there is a certain group of individuals (27%), who were shown to participate frequently (weekly) in sports and are committed to an active life-style. However, infrequent participants were shown to be the majority of the sample (41%). It is worth examining these figures from a health perspective. It has been suggested (e.g., Howley & Franks, 1992; Wankel & Berger, 1990) that in order to gain cardiorespiratory fitness benefits through sports and exercise participation an optimum frequency of three to four times per week is required. Consequently, those who participate at least once a month (17%) and especially those who participate less than once a month (24% of the sample) do not gain significant health benefits by their sport participation. The issue of the proportion of the participants who gain health benefits by their sport participation in Greece needs further investigation. Detailed measures of sport participation, in terms of the frequency, duration and intensity are necessary in order to clarify these issues.

Some clear patterns in terms of the most popular activities were revealed for males and females. Females were shown to be particularly attracted by health and fitness related activities, such as keep fit, aerobics, weight-training, jogging and dancing. In contrast, males were shown to be firstly attracted by the two team games, basketball and football, and secondly by fitness related activities, such as jogging, keep fit and weight-training. Walking was the only activity which was shown to be equally and highly popular for both the groups. Women's low participation rates in team and competitive activities might indicate that either women are not driven by competition related motives, or they do not participate in these activities because of the limited opportunities. Alexandris and Carroll (1997) provided support for the second argument. These authors investigated motives for taking part in recreational sports in Greece, and found that women were driven towards sport participation by a wide range of motives, included achievement / competition and intellectual related ones. Consequently, inadequate or inappropriate provision of sport programmes targeted women might be among the main reasons for women's' low participation rates in team and competitive activities. This is an issue with important practical implications for those planning, delivering and managing sport services in the area in which the research was conducted. If sport participation among women is to be increased, a wider range of sport programmes should be provided for them.

The decline in activities participation rates by age was expected. However, it is worth examining activities participation rates within the older age group (46-65). The most popular activities reported by this group were individual and unorganised activities, such as walking, football, jogging and cycling. Activities, such as aerobics and swimming, which have been suggested as very popular and appropriate for elderly individuals in other countries (e.g., England, Sports Council, 1994) had very low participation rates within the 46-65 group in the present study. It is worth noting that the annual participation rate of walking within the 46-65 group was 36%, much higher than the second most popular activity's participation rate within the same group (keep fit, 17%). These findings might reflect the poor and/ or inappropriate provision of sport services targeted older individuals (Alexandris and Carroll, 1997). Older individuals are a large and growing target group, who should not be overlooked by the sport organisations in Greece. Furthermore, these individuals have different needs and expectations, which should be carefully considered by those planning, delivering and managing the sport programmes in Greece.

Similar popular activities were stated by the primary, the secondary, and the university graduate groups. The high popularity of volleyball (41%) and basketball (41%) among the university students is a finding worth noting. Students were shown to be widely attracted by team and competitive sporting activities, which should be considered by both the university authorities and the sport clubs which target university students. This issue is particularly applicable to the private sport and fitness clubs, whose services are usually limited to indoor health and fitness related activities, such as aerobics, weight-training, martial arts and dancing sessions. The organisation of local leagues in sports, such as basketball and football (e.g., intramural games, games between the clubs) could be a possible suggestion.

Conclusions

Recreational sport participation and its effective promotion among the general population have been widely neglected by researchers in Greece. Based on a sample drawn from an urban area in Greece, the present study aimed to investigate patterns of recreational sport participation within the adult population. The study provided evidence that annual participation rates among the general population in Greece are low. Furthermore, the results indicated that only a small proportion of the population is committed to an active lifestyle, and participate frequently (weekly) in sports.

Some clear patterns were revealed in terms of the most popular sporting activities for the demographic groups used in the analysis. Females were shown to be widely attracted by health and fitness related activities, while males and especially young individuals were shown to be mainly attracted by competitive and team activities. Finally, walking was shown to be the most popular activity by far for the elderly individuals. These patterns might not reflect only individual preferences, but might also be related to the poor provision of sporting opportunities in the area in which the investigation was conducted.

Study Limitations and Future Research

A number of issues should be addressed which summarise the main limitations of the study, and indicate the need towards further research on aspects related to sport participation in Greece.

The sample size of the present study was not large. The sample size of five hundred individuals is subject to a certain degree of error {e.g., the confidence (95% chance) interval for a "finding" of 50% is +-4.5, for a finding of 30% / 70% is +-4.1 etc., see Veal, 1992}. Furthermore, the sample was not shown to be fully representative of the study population. Consequently, the results should be considered only as indicative and not representative of the study population. An example of a weighted sample, based on the 1991 census, was presented in an effort to indicate the possible effects of the biased sample on the results produced. Furthermore, the study collected information from an urban area, and the results could not be considered as representative of the population in Greece. Sport participation rates in rural areas might be lower. Further research is required by using both local and national samples of the Greek population. This could give the chance for more representative results, and for making comparisons with participation figures reported in other countries. Finally, regular (e.g., every year or every two years) collection of these kind of data could give the chance for time-series analysis, which could help to identify trends in sport participation. This is of particular importance for the sport and fitness industry.

As discussed previously, the study did not collect detailed information about the frequency of sport participation. Frequency was measured ordinally and for the year before the investigation. Both these methods have limitation which have already been discussed. Further research is required using more detailed measures of the frequency of participation (e.g., frequency, duration, intensity), which could give the chance to approach the issue of sport participation from a health perspective. Furthermore, participation rates based on the four week reference period should also be investigated.

It was argued that females' and older individuals' low participation rates in specific activities (e.g., swimming and aerobics) might be related to the inadequate or inappropriate provision of sport services. Unfortunately, there is no research available in Greece about the range and the quality of the sport services provided by both public and private sport organisations in order to support this argument. Future research on a variety of aspects related to the provision of recreational sport services (e.g., range and quality of programmes, competition between private and public organisations) is required in order to relate the present findings, and draw some conclusions.

References

Alexandris, K. & Carroll, B., (1997). Motives for Recreational Sport Participation in Greece. Implications for Planning and Provision of Sport Services. European Physical Education Review, 3(2), 129-143.

Alexandris, K. & Carroll, B. (1997). An Analysis of Leisure Constraints based on Different Recreational Sport Participation Levels: Results from a Study in Greece. Leisure Sciences, 19, 1-15.

Berger, B. (1996). Psychological Benefits of an Active Lifestyle: What we Know and What we Need to Know. Quest, 48, 330-353.

Boothby, J., Tungatt, F.M., Townsend, A., & Collins, M. (1981). A Sporting Chance? Family School and Environmental Influences on Taking Part in Sport. London: The Sports Council.

Census (1981). Information Provided by the Municipality of Larissa.

Chase, D., & Harada, M. (1984). Response Error in Self-reported Recreation Participation. Journal of Leisure Research, 16, 322-329.

Coalter, F., Dowers, S., & Baxter, M. (1995). The Impact of Social Class and Education on Sports Participation: Some Evidence from the General Household Survey. In Roberts, K., (Ed.), Leisure and Social Stratification (pp. 59-71). Leisure Studies Association, Publication No. 53.

Cushman, G., Veal, A.J., & Zuzanek, J. (1996). World Leisure Participation: Free Time in the Global Village. Wallingfod: Cab International.

Gambetta, C., DePauw, K. (1995). Attitudes toward Physical Activity among Germany and United States Senior Citizens. European Physical Education Review, 1, 155-162.

Gratton, C., & Taylor, P. (1985). Sport and Recreation. An Economic Analysis. London, New York: E & F.N. Spon.

Gratton, C., & Tice, A. (1994). Trends in Sports Participation in Britain. Leisure Studies, 13, 49-66.

Howard, D., & Crompton, J. (1984). Who are the Consumers of Public Park and Recreation Services? An Analysis of the Users and Non-users of three Municipal Leisure Service Organisations. JOURNAL of PARK aND RECREATION ADMINISTRATION, 2, 33-48.

Howley, E., & Franks, D. (1992). Health Fitness. Instructor's Handbook. Champaign, Illinois: Human Kinetics.

Kelly, R. (1980). Outdoor Recreation Participation: A Comparative Analysis. Leisure Sciences, 3, 129-154.

Kimiecik, J. & Lawson, H. (1996). Toward New Approaches for Exercise Behavior Change and Health Promotion. Quest, 48, 102-125.

Matheson, J. (1991). Participation in Sport. London: HMSO.

Nassis, P. (1996). An Analysis of Sports Policy in Greece (1980-92). In Collins, M. (Ed.), Leisure in Industrial and Post-Industrial Societies (pp. 31-41). Leisure Studies Association. Publication No. 49.

Sports Council (1994). Fifty Plus. All to Play for. A Manual for Organisers of Sport and Recreation for Older People. London: Sports Council.

Sport Development Department of Larissa, Greece (1995). Personal Contact with Sport Development Regional Officer.

Robinson, J., Godbey, G. (1993). Sport, Fitness and the Gender Gap. Leisure Sciences, 15, 291-307.

Samuel, N. (1996). France. In Cushman, G., Veal, A.J., & Zuzanek, J. (Eds.). World Leisure Participation: Free Time in the Global Village (pp. 77-106). Wallingfod: Cab International.

Stockdale, J., Wells, A., & Rall, M. (1996). Participation in Free-time Activities: A Comparison of London and New York. Leisure Studies, 15, 1-16.

Sharkey, B. (1990). Physiology of Fitness. Third Edition. Champaign, IL: Human Kinetics.

Shephard, R. (1986). Economic Benefits of Enhanced Fitness. Champaign, IL: Human Kinetics.

Shephard, R. (1995). Physical Activity, Fitness, and Health: The Current Consensus, Quest, 47, 288-303.

SPSS for Windows (1993). Base System User's Guide. Release 6.0. Chicago, Illinois: SPSS inc.

Taylor, P. (1995). Sinking or Swimming. Leisure Management, 15 (3), 38-43.

Veal, A. (1992). Research Methods for Leisure and Tourism. A Practical Guide. London: Longman. ILAM.

Veal, A., & Cushman, G. (1996). Leisure in Different Worlds: The Survey Evidence. In Collins, M. (Ed.), Leisure in Industrial and Post-Industrial Societies (pp. 403-412). Leisure Studies Association. Publication No. 49.

Wallace, N. (1996). Doing the Figures. Health Club Management, March, 19-20.

Wankel, L. & Berger, B. (1990). The Psychological and Social Benefits of Sport and Physical Activity. Journal of Leisure Research, 22, 167-182.