Myy.haaga-helia.fi
Spectators of Finnish baseball: comparing women’s and men’s games
Introduction
There are few studies that have compared the attendance of sport activities between men’s and women’s games or between genders. Most studies show that there are more male spectators than female (see Vuolle, Telama & Laakso 1986, Gantz & Wenner 1991, Zhang, Pease, Hui & Michaud 1995, White & Wilson 1999 or Thrane 2001). Women seem to favour women’s games and men favour men’s games respectively (Kahle, Duncan, Dalakas & Aiken 2001). The sociology of sport consumption has revealed that the motives for attending women’s games and men’s games differ. Typically, the aesthetics of the game or competition is more important for women’s team spectators and for female spectators (Ridinger & Funk 2006) while e.g. tracking statistics is more important for men (Fink, Trail & Anderson 2002).
The relationship between gender and active sport consumption i.e. participation in sport competitions or being a member in sport or gymnastic club reveals only minor differences in Finland. Both genders are as active but women seem to favour more clubs of commercial purposes (e.g. gym with aerobics) while men are more often members in sports associations that play games (Kansallinen liikuntatutkimus 2005-2006). Gymnastics at home and within a gymnastic association have been typically female while fishing and hunting have been male sport activities (Marin 1988).
The relationship between gender and passive sport consumption i.e. attendance at games has been less studied. If there are gender differences across games and if women spectators have different motives to attend sport activities, the factors explaining attendance should be different. Since men use more time in tracking statistics and reading about sports in daily newspapers (Dietz-Uhler, Harrick, End & Jacquemotte 2000), a team’s winning percentage or other previous performance measure of the team should be less important to explain women’s teams’ attendance figure. There are also differences between the importance of ticket pricing, friend influence and family involvement in women’s and men’s (basketball in the USA) games (Fink, Trail & Anderson 2002); hence the price elasticity of demand should differ. Women’s games should be more ticket price sensitive.
Sport activity in Finland
The motive of this study is to compare the attendance of women’s Finnish rule baseball[1] games with men’s games. The sociology of active sport consumption (recently Kansallinen liikuntatutkimus 2005-2006) reveals that baseball is not among the most popular sports in Finland (population 5,3 Million), the rank is between 30 and 35 with 16500 active players when e.g. floorball (223000 active players), football (160000) or ice hockey (90000) are more popular. Individual sport activity is, however, much more popular, e.g. light jogging (1840000), cycling (828000), skiing (747000) or swimming (578000) are exercised regularly and when possible. The ski season in southern Finland lasts few months from December to March and cycling during this winter season is almost impossible.
The Finnish data on the recent International Social Survey Programme (ISSP)[2] reveals that almost 40 % of the population never goes to see a sports activity (ice hockey, football, athletics, motor racing etc.), less than 8 % attends several times a month and the rest (i.e. more than 50 %) occasionally. The same survey also shows active sport consumption is more common (table 1: Sports consumption in Finland 2007).
(Table 1 about here)
There is a significant difference between genders so that males are more active in passive sports consumption (attendance) while females are more active exercisers. Almost regardless of the marital status men seem to attend more than women, single men as well as married or divorced men go more often to attend a game than women but there are no significant differences between genders living in cohabitation without marriage or widowed. Active and passive (attendance) sports consumption are only slightly positively correlated (Kendall’s tau = 0,054, n = 1314, sig. = 0,028). There is also a negative relationship between age and passive sports consumption (Spearman’s rho = -0,182, n = 1265, sig. = 0,000). For female, the relationship is somewhat stronger (Spearman’s rho = -0,193, n = 724, sig. = 0,000) than for male (Spearman’s rho = -0,179, n = 540, sig. = 0,000).
Another survey (Liikuntatutkimus 2005-2006, Sport Survey: Adult Population) on adult population sport consumption – both active and passive – in Finland was carried out few years ago[3]. The sample size was 5510. In this survey 44 % responded that they had not attended any sports event between February 2005 and January 2006. The most popular sports in terms of attendance were ice hockey (25.5%), football (16.9%), athletics (10.6%), skiing (6.5%) and Finnish rule baseball (5%). The correlation matrix (table 2) shows that Finnish rule baseball attendance is only moderately positively associated with other popular sports. The largest positive correlation is between ice hockey and football attendance. Attendance and income level (8 categories from the lowest to the highest) are not correlated (not reported here). However, correlation analysis will not reveal the association between attendance and incomes since other factors are not controlled. For instance in Canada age is negatively related to sport attendance and after controls (age) income and education are positive predictors of the percentage of both female and male respondents who have attended a sport event (White & Wilson 1999). High-income households are more likely to attend sport events also in Denmark, Norway and Sweden (Thrane 2001).
(Table 2 about here)
Recently the most popular sport event or sport league has been men’s ice hockey. During the regular series 2007-2008 there were 392 games with 1964626 spectators in the highest league (average 5012 per game). Theatre attendance was even larger but the number or performances were also higher: in 2007 almost 2.7 Million tickets sold in almost 13000 performances (average 207 per performance). At Finnish national opera there were 198 performances with 162555 tickets sold in 2007 (average 820 per performance). The premier league in male football with 182 games got 541612 spectators (average 2976 per game). The attendance in female ice hockey or football games has been substantially lower, during regular season 2007-2008 in the highest women’s ice hockey league of 6 teams with 60 games the total attendance was only 4022 (average 67 per game). Women’s highest football league with 10 teams and 110 games during the regular season in 2007 got 16384 spectators (average 149 per game).
Finnish rule baseball is more egalitarian in terms of attendance than ice hockey or football if both women’s and men’s games are taken into account. During the regular season 2007 excluding playoff games women’s 110 games got 55732 spectators (average 507 per game) and men’s 169 games got 235498 spectators (average 1393 per game). There is a big gap between women’s and men’s baseball games’ attendance but it is substantially lower than in ice hockey or football. Therefore baseball games are an ideal target to compare women’s and men’s games and their spectator figures. Since there is no regular series for skiing or athletics suitable statistics on attendance has not been collected.
About 5 percent of adult population has attended a baseball game according to Sport Survey: Adult Population (2005-2006). The share is somewhat higher for male (6.1 %) than for female (3.9%)[4]. The attendance is not correlated with family incomes or age (not reported here).
Motive differences between genders in sport consumption
The sociology of sport consumption has revealed that there are substantial motive differences between genders. A well-known classification is Sport Fan Motivation Scale (SFMS) by Wann (1995). There are eight motives: eustress (i.e. the need for positive stress), self-esteem (i.e. the desire to maintain a positive self-concept through team success), escape (i.e. sport as diversion from bored everyday life), entertainment, economic (i.e. gamble on the events), aesthetic (i.e. sport as an art), group affiliation (i.e. belongingness need), and family (i.e. opportunities to spend time with family). Wann conducted a quantitative examination with a 23-item Likert scale questionnaire. Using confirmatory factor analysis the above mentioned eight internally consistent, reliable and criterion valid motives were found. The original sample consisted of primarily of university college students. Several studies, however, confirmed the results (e.g. Wann, Schrader & Wilson 1999, Wann, Royalty & Rochelle 2002, Wann, Robinson, Dick % Gillentine 2003, Ridinger & Funk 2006, Wann, Grieve, Zapalac & Pease 2008 or Koo & Hardin 2008).
Wann, Schrader & Wilson (1999) found out those male spectators scored higher than female on the total SFMS and on the eustress, self-esteem, escape, economic and aesthetic as well as group affiliation subscales while female reported higher levels of family motivation. Any gender differences on entertainment subscale were not found. On the other hand, Kahle, Duncan, Dalakas & Aiken (2001), James & Ridinger (2002) and Ridinger & Funk (2006) show that women’s team spectators have a higher aesthetic factor loading than men’s team spectators.
Eustress, self-esteem and group affiliation motives were more associated with team and aggressive sport type (e.g. football, hockey) rather than individual and nonaggressive sport type. On the other hand aesthetic motive was associated with individual and nonaggressive sport type (e.g. figure skating, tennis). Wann, Schrader & Wilson (1999) also classify sport spectators as intrinsically or extrinsically oriented. Fans who enjoy sport because of its aesthetic and artistic movement (intrinsic) may not bother of their favorite team’s or individual’s poor performance since the aesthetic performance of the event is present regardless of the outcome. On the other hand extrinsic fans (self-esteem, economic motives) could find unpleasant to watch their favorite team’s games unless the team is victorious.
Self-respect and self-fulfillment are more associated with women’s team spectators (Kahle, Duncan, Dalakas & Aiken 2001) while self-indulgence is more men’s team spectators’ attribute. The opportunity to spend time with family or sense of belonging or socialization is attributes associated with women’s sport spectators (Kahle, Duncan, Dalakas & Aiken 2001 or Ridinger & Funk 2006). Females seem to be more sport fans for social reasons (Dietz-Uhler, Harrick, End & Jacquemotte 2000) while males are more likely to be fans because they play sports and want to acquire sport information (e.g. read sport pages in newspapers).
Sport events provide possibilities for groups (family, friends) to socialize and this enhances the overall experience (Hall & O’Mahony 2006). Sports venue on itself is an important factor in attendance. Spectators consider issues like size, seating comfort and access, also the cleanliness of the venue and the availability of car parking are important. The venue characteristics are more important for female than male (Hall & O’Mahony 2006). Women’s sport spectators emphasize more ticket pricing than men’s sport spectators (Fink, Trail & Anderson 2002). Melnick & Wann (2004) point out that females are more likely than males being influenced by school and parents while males are more likely being influenced by friends among Norwegian university students.
Motives explain a significant part of variance in attendance behavior but a great deal of variance across team’s attendance figures remains unexplained by motives. Several environmental variables have been found to influence attendance. Ticket prices, advertising efforts, size of the community, weather conditions, the performance of the favorite team (e.g. winning percentage) can explain variation across teams (e.g. Burdekin & Idson 1991, Coates & Harrison 2005, Garcia & Rodriguez 2002, McDonald & Rascher 2000). Distance between home town and visitor’s town has a significantly negative effect on attendance in MLB baseball (Knowles, Sherony and Haupert 1992).
However, since it has been found out that women seem to favour women’s games and men respectively men’s games, and since female spectators on average have different motives than male, the environmental factors, like ticket prices and team’s performance (winning percentage) should have different impact on the attendance of women’s games and men’s games. The gender differences in the sociology of sport consumption propose that female spectators are more intrinsically oriented and they care less about team’s performance (winning percentage) than male spectators (Aiken & Koch 2009).
Hypothesis 1: Since female spectators are probably more intrinsically oriented they care less about the team’s performance (winning percentage) than male spectators and winning percentage has less importance on the attendance of women’s games than men’s games.
Gender has a significant impact on the influence of family and friends on attendance decisions, and since ticket pricing is more important for women’s team spectators (Fink, Trail & Anderson 2002), the second hypothesis proposes that price elasticity is lower in men’s games (inelastic) than women’s games (elastic).
Hypothesis 2: Since female spectators favor women’s games and since the family of friends influence attendance decisions is more important for female, the ticket pricing is a more important factor in the attendance of women’s games and attendance is price elastic while in men’s games attendance is price inelastic.
Ross (2007) claims using a sample of 662 basketball (NBA) spectators that gender, educational level and household incomes are one the most important factor to distinguish spectators into two groups. Marital status, age and ethnicity are not separating factors. There are substantial differences between these two groups in commitment, concessions, team history, brand mark, organizational attributes, rivalry, personnel, stadium community, social interaction, success and team play (all considered as brand association dimensions). The most differentiating and important attributes to consumers’ preferences are related to the core of the sport product: the degree of sport popularity and physical contact during competition and there are substantial gender differences (Ferreira & Armstrong 2004).
Factors explaining passive Finnish baseball consumption
There is a wide literature on sport events’ attendance[5]. The literature explaining attendance in sport events, especially in the USA, is also wide starting with Demmert (1973) and Noll (1974). Conventional economic theory assumes that demand base measured as the incomes of the relevant market population and market size (population) should have an impact on attendance. Teams from bigger cities should have bigger attendance if the venue capacity allows it. Many teams are local monopolies with almost zero marginal costs of attendance. Hence maximizing profits equals maximizing revenues, and the outcome should be to set ticket prices high enough to ensure unitary price elasticity. Most studies still reveal that sporting events are priced in the inelastic range (Krautmann & Berri 2007).
A great majority of studies have focused on the series of matches or the regular league; hence it is natural that teams’ attendance is the explanatory variable and that teams from larger towns or cities have both bigger attendance and stadium capacity. Teams have a different demand base depending on home town’s population and several studies have confirmed the positive impact on attendance (e.g. Burdekin & Idson 1991, Depken 2000, Coates & Harrison 2005). The visitor’s home town population as well as the geographical distance between the towns is a significant factor to explain attendance (Knowles, Sherony & Haupert 1992, Suominen 2009). Game attractiveness (e.g. home team’s win/loss record, team history) is associated with attendance (Boyd and Boyd 1998, Burdekin and Idson 1991, Coates and Harrison 2005, Coates and Humphreys 2007, Depken 2000, Depken 2001, Kahane and Shmanske 1997, McDonald and Rascher 2000, Simmons 1996, Zhang, Lam & Connaughton 2003). Davis (2008) shows that the causation runs from win/loss record to attendance. Also when the season begins games have high interest since sport fans want to see new players (Wilson & Sim 1995).
The ticket price should be less important for men’s games than for women’s games. Attendance is rather inelastic in many (male) sports (e.g. Borland & Lye 1992, Carmichael, Millington & Simmons 1999, Depken 2001, Garcia & Rodriguez 2002, Coates & Harrison 2005, Coates & Humphreys 2007). The temperature or weather conditions have an impact on outdoor sports attendance (Baimbridge, Cameron & Dawson 1995, Jones, Schofield & Giles 2000, Garcia & Rodriguez 2002). The weekday effect is important, during weekends the attendance is bigger (Suominen 2009).
Based on the literature survey the following model explaining attendance can be formulated
[pic]
where ATTjk is the logarithm of attendance for each game when home team j has played against visitor k and Pricejk is the logarithm of the ticket price for that game and GSFjkm (game specific factors) describes interest towards the game that can be measured with home team’s j winning ratio as well as visitor’s k winning ratio in previous games as well as the market size (population). Time specific factors (TSFjk) are related with weather conditions and weekday effect.
Based on Coates & Harrison (2005) it is plausible to assume that interest towards the game is higher when the home team has won the previous games. Points-per-game from the beginning of the season is a suitable empirical measure for the winning ratio. The impact of visitor’s winning ratio is ambiguous since there is some evidence that competitive balance has a positive impact on the league (The USA, baseball) total attendance but a negative impact on team’s attendance (Schmidt & Berri 2001). Suominen (2009) shows with Finnish ice hockey attendance data that visitor’s winning ratio has a negative impact and for male ice hockey the form guide (winning percentage of the three last games) has a lower explanatory power than points per game from the beginning of season. Based on the first hypothesis the winning ratio should be less important to explain attendance of women’s games than for men’s games. The interest towards the game is positively related with both home town and visitor’s town population and negatively with geographical distance.
The second hypothesis claims that women’s games should be more price elastic than men’s games. Since baseball games are outdoor events, both temperature and a rainy weather should have an impact on attendance. Finally the weekday effect is important: during weekends the attendance should be higher. The sport venue is also important and since some teams’ home ground has a roof above the seating places, this roof variable should be taken into account.
The sample contains two years data. During the regular season in 2006 and 2007 there were 11 women teams and 12/13 men teams playing Finnish rule baseball in the highest league. Some of men’s games in 2006 were played in Helsinki where there is not any highest league team playing (i.e. both teams were visitors) and these six games have been excluded. Overall the number of women games is 220 in the sample (11 home teams x 10 visitors x 2 years) and the number of men games is 325 (2006: 156 and 2007: 169) in the sample – some teams met three times during the regular seasons 2006 and 2007.
(tables 3 and 4: about here)
Based on the coefficient of variation (tables 3 and 4) it seems that new teams in men’s league that played during the previous season on a lower league and therefore rose to the highest league have a bigger variation in attendance figures. Average attendance in women’s baseball during regular season 2006 and season 2007 are highly correlated and associated with population (table 5). Attendance variation also seems to be correlated with population during 2006 season but not during 2007 season. For men, the only significant correlation is between the average attendances of 2006 and 2007 (table 6). There is only one town with teams playing both in the highest league of baseball and of football: Kuopio. Football (and ice hockey or floorball) on the highest level is more urban than baseball since a larger amount of highest league football (and ice hockey or floorball) teams were (2006 or 2007) from the largest cities (Appendix, table 2). The urbanization rate might have an impact on attendance since baseball seems to be played in less urban towns.
(tables 5 and 6 about here)
Estimation techniques and results
Since the data has both time-series and cross-sectional (different home teams) dimension conventional regression analysis might be problematic. Panel data analysis enables regression analysis with both time-series and cross-sectional dimension. Panel data can have group effects (teams), time effects or both. Panel data models estimate fixed and/or random effects models using dummy variables. The core difference between fixed and random effect models lies in the role of dummies. If dummies are considered as a part of the intercept, it is a fixed effect model. In a random effect model, the dummies act as an error term. The fixed effect model examines team differences in intercepts, assuming the same slopes and constant variance across teams. Fixed effect models use least square dummy variable (LSDV), within effect, and between effect estimation methods. Thus, ordinary least squares (OLS) regressions with dummies, in fact, are fixed effect models. The random effect model, by contrast, estimates variance components for groups and error, assuming the same intercept and slopes. The difference among groups (or time periods) lies in the variance of the error term. This model is estimated by generalized least squares (GLS) when the variance structure among genres, is known. The feasible generalized least squares (FGLS) method is used to estimate the variance structure the variance structure among genres is not known.
Least square dummy variable (LSDV) model, however, becomes problematic when there are many groups or subjects in the panel data. If the total number of periods is fixed and the total number of observations is vast, only the coefficients of regressors are consistent. The coefficients of dummy variables are not consistent since the number of these parameters increases as N increases (Greene 2008, 197). This is so called the incidental parameter problem. Too many dummy variables may weaken the model for adequately powerful statistical tests. Under this circumstance, LSDV is useless, and another method might be used: the within effect model which does not use dummy variables, but uses deviations from group means.
Furthermore, if the venue has a covered stand it should have an impact on attendance but it might also affect the ticket pricing. Especially women spectators consider venue characteristics very important (Hall & O’Mahony 2006). Both ticket pricing and attendance are determined simultaneously and a system estimation method should be used. The equations might be connected not because they interact but because their error terms are related. If the explanatory variables are correlated with error terms, an instrumental variable (IV) method should be used. The problematic variable is replaced with an instrumental variable that is uncorrelated with the residual. A method of moments or the generalised method of moments (GMM) provides consistent estimators. One computational method is Two-Stage Least Squares (2SLS). At the first stage, each endogenous variable (attendance and ticket price) in the equation of interest is regressed on all of the exogenous variables in the model, including both exogenous covariates in the equation of interest and the excluded instruments. The predicted values from these regressions are obtained. At the second stage, the regression of interest is estimated as usual; except that in this stage each endogenous covariate is replaced with the predicted values from its first stage model from the first stage (see Kennedy 1998, 165). If only the error terms are related and equations are otherwise unrelated, a seemingly unrelated regression (SUR) improves the efficiency of estimation. Three-stage Least Squares (3SLS) is the systems counterpart of 2SLS and it is asymptotically more efficient than 2SLS. If the errors in the different equations are uncorrelated, 3SLS reduces to 2SLS.
Most Finnish baseball players, especially within the female league are not professionals but some compensation is paid to players. The compensation for the lost income to unprofessional players due to normal daily work absence is paid when the team plays guest game far away. This payroll must be financed partly by ticket revenue. Therefore the payroll has an impact on ticket prices, as well as the home town population and a dummy variable for the covered stand. In summary the structural model is
[pic]
[pic]
Descriptive statistics and correlation matrix of female baseball variables are shown in table 7
(Table 7 about here)
Attendance (ATT) seems to correlate positively with home town population (PopH), the urbanization degree of the town (urban). Ticket prices (Price) are higher is larger towns (PopH) and both are positively associated with a stadium having a covered spectator stand (Roof). Better teams in terms of the winning percentage (points per game, PPGH) seem to be located in bigger, more urban towns. The previous variables are associated with the payroll. Hence bigger compensation or wage to players is paid in larger towns and these teams seem to attract bigger attendance. Weather conditions in terms of temperature (Temp), a rainy day (Rain) or a windy day (Wind) are associated and they seem to correlate with attendance. The attendance is significantly higher on Fridays and lower on Saturdays. When the geographical distance between the home team and the visitor is big, the games are more often played during weekends (Saturdays or Sundays) and less on Tuesdays and Wednesdays. Teams from bigger towns seem to have less home games on Saturdays and this is probably the reason for lower attendance figures on Saturdays. Most female games were played on Wednesdays (37%) or Sundays (34%) during the regular seasons 2006 and 2007.
The estimation results – both OLS and random effects model (panel) – indicate that the ticket price is not significant variable and it has the incorrect sign. Home town population has a positive effect while urbanization has a negative effect on attendance. Geographical distance between the home team and the visitor has the correct negative sign. The temperature and a rainy day (dummy) effects are significant. A higher temperature, one degree in Celsius, increases attendance by roughly 1½ percent while rainy weather decreases attendance by roughly 25 – 29 percent. Having a covered spectator stand (Roof) increases attendance by 30 - 32 percent. The payroll has an impact: a higher payroll results in higher attendance. The payroll of the team and the overall budget of the team are correlated (r = 0,734) but there are no data about marketing costs or marketing mix, hence the payroll is proxy for marketing expenditure. During Saturdays, the attendance is significantly lower. The winning percentage has no influence on attendance. However, these estimation results are unsatisfactory since the ticket price is not significant. The SURE estimation is more satisfactory since the ticket price has a negative and significant coefficient. An important observation is that the hypothesis concerning price elasticity for female games is supported. The price elasticity is roughly -5 or -6. The previous winning percentage is not significant which supports the first hypothesis. The ticket price is determined by the payroll of the team, the urbanization degree of the town and the roof above the seating places. However, the payroll has a negative coefficient which is not plausible. The payroll and the urbanization degree are positively correlated and these are associated with having a covered spectator stand. The marginal effect of the payroll is negative since the teams will charge a low price to attract spectators and the payroll costs are sunk. The impact of having a covered spectator stand (roof) on the attendance seems to be very large, i.e. more than 60 percent.
(tables 8 and 9 about here)
Men’s games have had a bigger attendance than women’s. During the regular seasons 2006 and 2007 the figures were 1374 vs. 467 and the average ticket price was higher: 9€ vs. 6.82€. Men have had more stadiums with a covered spectator stand: 48% vs. 26% and the payroll of the team was higher but men have had also more games per season. Otherwise the explanatory covariates are similar to men’s and women’s games.
Majority of men’s games have been played on Tuesdays (24.6%), Thursdays (31.4%) or Sundays (33.8%). On Tuesday, the visitor has come from near as the negative relation between the Tuesday dummy and distance reveals. The home town population is positively correlated with the Tuesday dummy and negatively with the Friday dummy. The team with a bigger payroll seems to have played more home games on Thursdays. However, the connections of the days of the week to other variables are no more than slight.
The ticket price is correlated with the payroll and the winning percentage. The payroll is negatively correlated with the population of the town and positively with the winning percentage of the team. These correlations indicate that the best teams come from less urban areas, from the countryside where the spectators have less supply of leisure time activities. In bigger towns, there is more often a covered spectator stand. Hence it seems that the urban, city teams invest more to the venue quality and the rural, municipality teams to the player quality. According to expectations, the rainy weather and the temperature are (negatively) correlated and these are correlated with the spectator number.
(Table 10 about here)
The panel estimation results indicate that attendance is price sensitive only with fixed effects model, the price elasticity is roughly -1,5 but the significance level is only 10 percent. Home town population is not significant whereas the visitor’s town population has a negative coefficient. Since town population and the payroll are negatively correlated, teams from smaller towns are better in terms of player quality and the visitor’s town population variable might capture the visitor team’s quality. Therefore it measures the visitor’s performance (points per game, LogPPGV) which is not significant per se. Home team’s performance (points per game, LogPPGH) has the expected positive coefficient and the geographical distance between home town and visitor town has the negative coefficient which verifies the presumption. Temperature matters but the significance of the coefficient is only about 10 percent. A rainy weather reduces attendance by 17 - 18 percent. The effect of having a covered spectator stand is not significant. The SUR estimation is not satisfactory since the ticket price has the incorrect sign. However, the SUR estimation indicates that the ticket price is determined by the size of the town, the payroll of the team and negatively by having a covered spectator stand. In bigger towns or cities there is more often a covered spectator stand and the teams on average are not among the best measured by the winning percentage. The investment cost of the roof is sunk and that does not seem to be taken into account when the ticket pricing decisions are made.
(table 11 about here)
The impact of ticket price and winning percentage on attendance
The results indicate that women’s games are more sensitive to ticket prices. The price elasticity falls between ranges -5.3 … -5.9 while the corresponding elasticity of attendance of men’s games is about -1,5 (Fixed effects panel, significance level 10 percent). The average attendance is lower for women’s games than for men’s games regardless of the size of the community. Women’s games seem to attract more spectators from more urban cities and the venue characteristics, i.e. a covered spectator stand is more important for the spectators of women’s games. The male baseball is less urban and the venue quality is not so important.
The weather conditions, both the temperature and the possibility of rain have a bigger impact on women’s games’ attendance. One degree higher (Celsius) temperature increases women’s baseball attendance by 1½ percent while the corresponding impact is less than 1 percent for men’s baseball. The same is true with a rainy weather: for women’s baseball the decrease is 25 – 29 percent while form men’s baseball the decrease is 18 percent.
Team’s performance, i.e. the winning percentage is important factor in explaining men’s baseball attendance, however, only with fixed effects panel data. The SUR estimations are unsatisfactory since they do not reveal this effect. For women’s games this winning percentage factor is not significant.
Conclusions
Both hypotheses are valid. H1: Winning percentage is important explanatory variable for men’s baseball games while it is not for women’s games. The second hypothesis, H2: “women’s games are more price sensitive than men’s” is confirmed. The results, however, are not robust since the fixed effects panel data model seems to be suitable for explaining men’s attendance figures while it is unsatisfactory to explain women’s games. On the contrary seemingly unrelated regression (SUR) is more suitable for explaining women’s games and it is not satisfactory at all to explain men’s games.
Home town population has a positive effect on female game attendance but no effect on male game attendance. Most games have been played on Tuesdays (men), Wednesdays (women), Thursdays (men) and Sundays (women and men). There seems to be a Saturday effect: then attendance is lower but there not many games played on Saturdays. Otherwise the weekday effect is insignificant. Men’s games have a bigger attendance than women’s games and the ticket price is higher for men’s games. The estimation results indicate that female baseball league attendance is more price sensitive, therefore the lower ticket price in female baseball is reasonable.
The correlation matrix in table 2 (Attendance popularity and correlation among adult population in Finland, 2005-2006) reveals that baseball attendance is practically not correlated with other popular sports attendance. There is strong evidence that population is highly correlated with ice hockey or football attendance but since the baseball spectators are not the same people the population is not a significant explanatory variable in men’s baseball attendance. Highest league football and baseball are practically not played in same cities and therefore these events are not substitutes.
The venue or the stadium characteristics seem to be not important for men and therefore teams in male league should focus on the team quality in order to win more games and attract a bigger attendance. However, this is not the case in female league in which the venue characteristics – like having a covered spectator stand – are important to spectators. On average the team Porin Pesäkarhut (located in Pori: population about 76000, average attendance was 1131 in 2006 and1143 in 2007) has the largest spectator number and the stadium has a covered spectator stand while the second and third teams in terms of attendance, i.e. Jyväskylän kirittäret (Jyväskylä, population about 85000, average was attendance 573 in 2006 and 598 in 2007) does not have a covered spectator stand and Lappeenrannan Pesä-Ysit (Lappeenranta, population about 60000. average attendance was 498 in 2006 and 745 in 2007) has a covered spectator stand. The estimation results indicate that having a covered spectator stand increases attendance at least by 30 percent.
References
Aiken, K. Damon & Eric C. Koch (2009): A Conjoint Approach Investigating Factors in Initial Team Preference Formation. Sport Marketing Quarterly, 18, 81-91
Baimbridge, Mark, Samuel Cameron & Peter Dawson (1996) Satellite television and the demand for football: a whole new ball game? Scottish Journal of Political Economy, 43, 317-333
Borland, Jeffery & Robert Macdonald (2003): Demand for sport. Oxford Review of Economic Policy, 19, 478-502
Boyd, David W. & Laura A. Boyd (1998) The Home Field Advantage: Implications for the Pricing of Tickets to Professional Team Sporting Events. Journal of Economics and Finance, 22, 169-179
Breusch, T. and Pagan, A. (1980): The LM Test and Its Applications to Model Specification in Econometrics. Review of Economic Studies 47: 239-254
Burdekin, Richard C.K. & Todd L. Idson (1991) Customer preferences, attendance and the racial structure of professional basketball teams. Applied Economics, 23, 179-186
Carmichael, Fiona, Janet Millington & Roberts Simmons (1999) Elasticity of demand for Rugby League attendance and the impact of BskyB. Applied Economics Letters, 6, 797-800
Coates, Dennis & Thane Harrison (2005) Baseball strikes and the Demand of Attendance. Journal of Sports Economics, 6, 282-302
Coates, Dennis & Brad R. Humphreys (2007) Ticket Prices, Concessions and Attendance at Professional Sporting Events. International Journal of Sport Finance, 2, 161-170
Davis, Michael D. (2008): The Interaction between Baseball Attendance and Winning Percentage: A VAR Analysis. International Journal of Sport Finance, 3, 58-73
Depken, Craig A. II (2000): Fan Loyalty and Stadium Funding in Professional Baseball. Journal of Sports Economics, 1, 124-138
Depken, Craig A. II (2001): Fan Loyalty in Professional Sports: An Extention of the National Football League. Journal of Sports Economics, 2, 275-284
Dietz-Uhler, Beth, Elizabeth A. Harrick, Christian End & Lindy Jacquemotte (2000): Sex differences in sport fan behaviour and reasons for being a sport fan. Journal of Sport Behavior, 12, 219-231
Ferreira, Mauricio & Ketra L. Armstrong (2004): An Exploratory Examination of Attributes Influencing Students’ Decisions to Attend College Sport Events. Sport Marketing Quarterly, 13, 194-208
Fink, Janet S., Galen T. Trail & Dean F. Anderson (2002): Environmental factors associated with spectator attendance and sport consumption behavior: Gender and team differences. Sport Marketing Quarterly, 11, 8- 19
Gantz, W & L.A. Wenner (1991): Men, women and sports: Audience experiences and effects. Journal of Broadcasting and Electronic Media, 35, 233-243
García, Jaume & Plácido Rodríguez (2002) The Determinants of Football Match Attendance Revisited: Empirical Evidence From the Spanish Football League. Journal of Sports Economics, 3, 18-38
Greene, William H. (2008): Econometric Analysis. 6th Edition. Pearson International Edition.
Hall, John & Barry O’Mahony (2006): An empirical analysis of gender differences in sports attendance motives. International Journal of Sports Marketing & Sponsorship, 7, 334-346
Hausman, J. (1978): Specification Test in Econometrics. Econometrica 46: 1251-1271
James, Jeffrey D. & Lynn L. Ridinger (2002): Female and male sport fans: A comparison of sport consumption motives. Journal of Sport Behavior, 25, 260-278
Kansallinen liikuntatutkimus 2005-2006. SLU:n julkaisusarja 5/06. 2006
Kahane, Leo & Stephen Shmanske (1997) Team roster turnover and attendance in major league football. Applied Economics, 29, 425-431
Kahle, Lynn, Marc Duncan, Vassilis Dalakas & Damon Aiken (2001): The social values of fans for men’s versus women’s university basketball. Sport Marketing Quarterly, 10, 156-162
Kennedy, Peter (1998): A Guide to Econometrics. 4th Edition. Blackwell Publishers
Koo, Gi-Yong & Robin Hardin (2008): Difference in interrelationship between spectators’ motives and behavioral intentions based on emotional attachment. Sport Marketing Quarterly, 17, 30-43
Knowles, Glenn, Keith Sherony 6 Mike Haupert (1992) The demand for major league baseball: a test of the uncertainty of the outcome hypothesis. The American Economist, 36, no 2, 72-80
Marin, Marjatta (1988): Gender-Differences in sport movement in Finland. International Review for Sociology of Sport, 23, 345-359
McDonald, Mark & Daniel Rascher (2000) Does Bat Day Make Cents? The Effect of Promotions on the Demand for Major League Baseball. Journal of Sport Management, 14, 8-27
Melnick, Merrill J. & Daniel L. Wann (2004): Sport Fandom Influences, Interests, and Behaviors among Norwegian University Students. International Sports Journal, 8, 1-13
Park, Hun Myoung: Linear Regression Models for Panel Data Using SAS, STATA, LIMDEP, and SPSS. accessed 5th February 2008
Ridinger, Lynn L. & Daniel C. Funk (2006): Looking at gender differences through the lens of sport spectators. Sport Marketing Quarterly, 15, 155-166
Ross, Stephen D. (2007): Segmenting Sport Fans Using Brand Associations: A Cluster Analysis. Sport Marketing Quarterly, 16, 15-24
Schmidt, Martin B. & David J. Berri (2001) Competitive Balance and Attendance: The Case of Major League Baseball. Journal of Sports Economics, 2, 145-167
Simmons, Robert (1996) The demand for English league football: a club-level analysis. Applied Economics, 28, 139-155
Simmons, Rob (2006): The demand for spectator sports. In Handbook on the Economics of Sport, edited by Wladimir Andreff & Stefan Szymanski, 77-89, Edward Elgar
Suominen, Seppo (2009): Demand for Ice Hockey, The Factors Explaining Attendance of Ice Hockey Games in Finland. Paper presented at First European Conference in Sports Economics 14.-15.9.2009, Centre D’Economie de la Sorbonne, Université Paris 1, Panthéon-Sorbonne
Thrane, Christer (2001): Sport spectatorship in Scandinavia: A class phenomenon? International Review for the Sociology of Sport, 36, 149-163
Wann, Daniel L. (1995): Preliminary validation of the sport fan motivation scale. Journal of Sport & Social Issues, 19, 29-42
Wann, Daniel L., Frederick G. Grieve, Ryan K. Zapalac & Dale G. Pease (2008): Motivational Profiles of Sport Fans of Different Sports. Sport Marketing Quarterly, 17, 6-19
Wann, Daniel L., Matthwe J. Robinson, Ronald J. Dick & Andrew J. Gillentine (2003): Motives and Points of Attachment: Fans Versus Spectators in Intercollegiate Athletics. Sport Marketing Quarterly, 12, 217-227
Wann, Daniel L., Joel L. Royalty & Al R. Rochelle (2002): Using motivation and team identification to predict sport fans’ emotional responses to team performance. Journal of Sport Behavior, 25, 207-216
Wann, Daniel L., Michael P. Schrader & Anthony M. Wilson (1999): Sport fan motivation: Questionnaire validation, comparisons by sport, and relationship to athletic motivation. Journal of Sport Behavior, 22, 114-139
White, Philip & Brian Wilson (1999): Distinctions in the stands: An investigation of Bourdieu’s ‘Habitus’, socioeconomic status and sport spectatorship in Canada. International Review for the Sociology of Sport, 34, 245-264
Wilson, Peter & Benson Sim (1995): The Demand for Semi-Pro League Football in Malaysia 1989-91: A Panel Data Approach. Applied Economics, 27, 131-138
Vuolle, P, R. Telama & L. Laakso (1986): Näin suomalaiset liikkuvat. Liikunnan ja kansanterveyden julkaisuja, 50. Helsinki
Zhang, James J, Eddie T.C. Lam & Daniel P. Connaughton (2003): General Market Demand Variables Associated with Professional Sport Consumption. International Journal of Sports Marketing & Sponsorship, 5, 33-55
Zhang, J., D. Pease, S. Hui & T. Michaud (1995) : Variables affecting the spectator decision to attend NBA games. Sport Marketing Quarterly, 4, 29-39
Eurobarometer 67.1 February – March 2007.
ISSP 2007 : vapaa-aika ja urheilu : Suomen aineisto [elektroninen aineisto]. FSD2330, versio 1.0 (2008-03-28). Blom, Raimo (Tampereen yliopisto. Sosiologian ja sosiaalipsykologian laitos) & Melin, Harri (Turun yliopisto. Sosiologian laitos) & Tanskanen, Eero (Tilastokeskus. Haastattelu- ja tutkimuspalvelut) [tekijä(t)]. Tampere : Yhteiskuntatieteellinen tietoarkisto [jakaja], 2008.
Liikuntatutkimus 2005-2006 : aikuisväestö [elektroninen aineisto]. FSD2233, versio 1.0 (2007-01-29). Helsingin kaupunki, Nuori Suomi, Opetusministeriö, Suomen Kuntoliikuntaliitto, Suomen Liikunta ja Urheilu, Suomen Olympiakomitea & Pehkonen, Juhani (Suomen Gallup) [tekijät]. Tampere : Yhteiskuntatieteellinen tietoarkisto [jakaja], 2007.
Rules of Finnish baseball (quoted 23.10.2009):
| |Daily |Several times a |Several times a |Occasionally |Never |Total, n |
| | |week |month | | | |
|How often do you attend a |4 (0,3%) |17 (1,3%) |82 (6,2%) |691 (52,3%) |526 (39,8%) |1320 |
|sports activity? | | | | | | |
|How often do you attend a |1 (0,1%) |5 (0,7%) |38 (5,1%) |327 (44,2%) |369 (49,9%) |740 |
|sports activity? (Female)* | | | | | | |
|How often do you attend a |3 (0,5%) |12 (2,1%) |43 (7,6%) |358 (63,3%) |150 (26,5%) |566 |
|sports activity? (Male) | | | | | | |
|How often do you exercise |301 (22,6%) |546 (41,0%) |272 (20,5%) |183 (13,8%) |28 (2,1%) |1330 |
|sports? | | | | | | |
|How often do you exercise |189 (25,5%) |308 (41,5%) |144 (19,4%) |92 (12,4%) |9 (1,2%) |742 |
|sports? (Female)** | | | | | | |
|How often do you exercise |106 (18,5%) |230 (40,1%) |127 (22,2%) |91 (15,9%) |19 (3,3%) |573 |
|sports? (Male) | | | | | | |
|Table 1: Sports consumption in Finland 2007, International Social Survey Programme, own calculations |
|*Significantly different between genders, Mann-Whitley U-test, (z= -8,430, sig. = 0,000) |
|**Significantly different between genders, Mann-Whitney U-test (z=-3,858, sig=0,000) |
| |Popularity |Ice Hockey |Football |Athletics |Skiing |F Rule Baseball |
|Ice Hockey |25,5% |1 | | | | |
| |F: 14,6% | | | | | |
| |M: 36,4% | | | | | |
|Football |16,9% |0,323 (0.000) |1 | | | |
| |F: 11,0% |F: 0,353 (0.000) | | | | |
| |M: 22.8% |M:0,193(0,000) | | | | |
|Athletics |10,6% |0,093 (0,000) |0,123 (0,000) |1 | | |
| |F: 9,9% |F: 0,133 (0,000) |F: 0,156 (0,000) | | | |
| |M: 11,3% |M: 0,031 (0,108) |M:0,074 (0,000) | | | |
|Skiing |6,5% |0,009 (0,517) |0,022 (0,110) |0,147 (0,000) |1 | |
| |F: 6,3% |F: 0,002 (0,909) |F:0,019 (0,315) |F:0,150 (0,000) | | |
| |M: 6,6% |M: 0,015 (0,431) |M: 0,024 (0,216) |M: 0,143 (0,000) | | |
|F Rule Baseball |5,0% |0,098 (0,000) |0,056 (0,000) |0,056 (0,000) |0,014 (0,295) |1 |
| |F: 3,9% |F: 0,085 (0,000) |F: 0,050 (0,009) |F: 0,053 (0,006) |F: 0,024 (0,212) | |
| |M: 6,1% |M: 0,096 (0,000) |M: 0,049 (0,010) |M: 0,058 (0,002) |M: 0,001 (0,942) | |
|Table 2: Attendance popularity and correlation among adult population in Finland, 2005-2006, n = 5510, significance in parenthesis, female n = 2754,|
|male n = 2756 |
|Women league team |Home game |Variation of home game |Home game |Variation of home game attendance: |
| |average |attendance: |average |min – max (std), |
| |attendance |min – max (std), |attendance |coefficient of variation |
| |(2006) |coefficient of variation |(2007) |(2007) |
| | |(2006) | | |
|Hamina |297 |385 – 214 (47,4) |292 |447 – 140 (98,9) |
| | |0,16 | |0,34 |
|Jyväskylä |573 |1489 – 268 (353,8) |598 |993 – 189 (296,0) |
| | |0,62 | |0,50 |
|Lappeenranta |498 |1364 – 327 (308,1) |745 |2268 – 367 (550,7) |
| | |0,62 | |0,74 |
|Lapua |406 |553 – 289 (76,4) |540 |782 – 337 (169,3) |
| | |0,19 | |0,31 |
|Peto-Jussit |323 |545 – 226 (118,8) |438 |647 – 237 (152,1) |
| | |0,37 | |0,35 |
|Pori |1131 |1937 – 365 (426,8) |1143 |1383 – 990 (138,9) |
| | |0,38 | |0,12 |
|Rauma |474 |704 – 267 (119,0) |411 |506 – 242 (77,2) |
| | |0,25 | |0,19 |
|Siipe |313 |381 – 192 (62,5) |401 |913 – 231 (240,8) |
| | |0,20 | |0,60 |
|Sotkamo |174 |245 - 140 (35,4) |- |- |
| | |0,20 | | |
|Tyrnävä |343 |517 – 248 (73,8) |391 |678 – 267 (122,3) |
| | |0,22 | |0,31 |
|Viinijärvi |- |- |353 |513 – 186 (125,2) |
| | | | |0,36 |
|Ylihärmä |168 |243 – 120 (43,2) |263 |385 – 162 (67,2) |
| | |0,26 | |0,25 |
|Table 3: Women’s baseball regular season 2006 and 2007 average attendance statistics, source: superpesis.fi |
|Men league team |Home game |Variation of home game |Home game |Variation of home game attendance: |
| |average |attendance: |average |min – max (std), |
| |attendance |min – max (std), |attendance |coefficient of variation |
| |(2006) |coefficient of variation |(2007) |(2007) |
| | |(2006) | | |
|Hyvinkää |1385 |2957 – 657 (543,2) |1878 |3629 – 1112 (804,6) |
| | |0,39 | |0,43 |
|Imatra |691 |1039 – 475 (182,4) |- | |
| | |0,26 | | |
|Joensuu |1235 |2184 – 512 (399,1) |1396 |2002 – 436 (454,9) |
| | |0,32 | |0,33 |
|Jyväskylä |- | |951 |2316 – 597 (433,0) |
| | | | |0,46 |
|Kitee |1690 |2735 – 1014 (521,7) |1669 |2139 – 1013 (398,3) |
| | |0,31 | |0,24 |
|Koskenkorva |987 |2242 – 658 (466,2) |1161 |1589 – 690 (263,0) |
| | |0,47 | |0,23 |
|Kouvola |1121 |2190 – 812 (359,5) |1259 |1922 – 855 (305,4) |
| | |0,32 | |0,24 |
|Kuopio |1203 |2411 – 782 (441,6) |1413 |2577 – 679 (502,5) |
| | |0,37 | |0,36 |
|Nurmo |1204 |1545 – 816 (242,0) |1152 |1864 – 810 (281,9) |
| | |0,20 | |0,24 |
|Oulu |1714 |3066 – 854 (617,2) |1292 |2010 – 740 (381,3) |
| | |0,36 | |0,30 |
|Pattijoki |1064 |1536 – 887 (189,6) |1277 |2007 – 888 (333,1) |
| | |0,18 | |0,26 |
|Seinäjoki |- | |1101 |2812 – 569 (661,5) |
| | | | |0,60 |
|Sotkamo |1724 |3161 – 1036 (683,7) |1756 |2961 – 1100 (515,9) |
| | |0,40 | |0,29 |
|Vimpeli |1659 |2916 – 857 (551,9) |1811 |3021 – 980 (610,7) |
| | |0,33 | |0,34 |
|Table 4: Men’s baseball regular season 2006 and 2007 average attendance statistics, source: superpesis.fi |
| |Average attendance, |Average attendance, |Population 2006 |Coefficient of |Coefficient of |
| |Female 2006 |Female 2007 | |Variation, Female |Variation, Female |
| | | | |attendance 2006 |attendance 2007 |
|Average attendance, |1 | | | | |
|Female 2006 | | | | | |
|Average attendance, |0,946** |1 | | | |
|Female 2007 | | | | | |
|Population 2006 |0,782** |0,768** |1 | | |
|Coefficient of |0,432 |0,520 |0,824** |1 | |
|Variation, Female | | | | | |
|attendance 2006 | | | | | |
|Coefficient of |-0,270 |-0,050 |0,173 |0,497 |1 |
|Variation, Female | | | | | |
|attendance 2007 | | | | | |
|Table 5: Correlation matrix of selected variables, female baseball |
| |Average attendance, |Average attendance, |Population 2006 |Coefficient of |Coefficient of |
| |Male 2006 |Male 2007 | |Variation, Male |Variation, Male |
| | | | |attendance 2006 |attendance 2007 |
|Average attendance, |1 | | | | |
|Male 2006 | | | | | |
|Average attendance, |0,669* |1 | | | |
|Male 2007 | | | | | |
|Population 2006 |0,128 |-0,308 |1 | | |
|Coefficient of |0,281 |0,279 |0,182 |1 | |
|Variation, Male | | | | | |
|attendance 2006 | | | | | |
|Coefficient of |0,234 |-0,134 |0,296 |0,281 |1 |
|Variation, Male | | | | | |
|attendance 2007 | | | | | |
|Table 6: Correlation matrix of selected variables, male baseball |
|Variable |
|Model |OLS without group dummy variables |Random effects model (REM) | | |
| | | |OLS |REM |
|LogPrice |0,365 |0,555 |0,426 |0,606 |
| |(0,251) |(0,330) |(0,246) |(0,326) |
|LogDist |-0,110 |-0,172 |-0,118 |-0,166 |
| |(0,043)** |(0,038)*** |(0,038)*** |(0,034)*** |
|LogPopH |0,329 |0,364 |0,345 |0,386 |
| |(0,055)*** |(0,111)*** |(0,054)*** |(0,108)*** |
|LogPopV |0,023 |0,030 |0,031 |0,037 |
| |(0,026) |(0,022) |(0,025) |(0,022) |
|LogPPGH |0,009 |0,006 |0,006 |-0,000 |
| |(0,019) |(0,018) |(0,019) |(0,017) |
|LogPPGV |0,018 |0,012 |0,016 |0,009 |
| |(0,016) |(0,014) |(0,016) |(0,014) |
|Temp |0,013 |0,013 |0,013 |0,016 |
| |(0,006)* |(0,005)** |(0,005)* |(0,005)** |
|Rain |-0,289 |-0,247 |-0,297 |-0,251 |
| |(0,075)*** |(0,065)*** |(0,074)*** |(0,064)*** |
|Wind |-0,010 |-0,041 | | |
| |(0,056) |(0,049) | | |
|Roof |0,327 |0,300 |0,327 |0,303 |
| |(0,062)*** |(0,132)** |(0,061)*** |(0,130)** |
|LogUrban |-0,795 |-1,102 |-0,919 |-1,278 |
| |(0,331)* |(0,613) |(0,324)** |(0,599)* |
|LogPayroll |0,285 |0,371 |0,304 |0,416 |
| |(0,065)*** |(0,106)*** |(0,063)*** |(0,103)*** |
|Monday |0,044 |0,137 | | |
| |(0,207) |(0,182) | | |
|Tuesday |0,092 |0,055 | | |
| |(0,103) |(0,088) | | |
|Wednesday |-0,008 |-0,057 | | |
| |(0,062) |(0,054) | | |
|Thursday |0,122 |0,132 | | |
| |(0,092) |(0,079) | | |
|Friday |0,182 |-0,005 | | |
| |(0,133) |(0,119) | | |
|Saturday |-0,329 |-0,292 |-0,348 |-0,289 |
| |(0,09)*** |(0,086)*** |(0,093)*** |(0,082)*** |
|Constant |2,57 |2,59 |2,61 |2,39 |
| |(0,989)** |(1,62) |(0,961)** |(1,58) |
|Adjusted R-sq |0,627 | | |
|F-test |21,46*** | | | | | |
|Diagnostic LL |235,88*** | | | | | |
| |Test statistics for the classical model | | |Test statistics for the classical model |
| |Constant term only |Log likelihood | | |Constant term only |Log likelihood |
| |(1) |= -185,70 | | |(1) |= -185,70 |
| |X-variables only (3)|LL = -67,72 | | |X-variables only (3)|LL = -70,31 |
| |Hypothesis tests | | | |Hypothesis tests | |
| |(3) vs. (1) |235,88*** |(REM) vs. (3) |103,20*** |(3) vs. (1) |230,79*** |
| | | |Baltagi-Li form |89,91 | | |
| | | | | |(REM) vs. (3) |106,02 |
| | | | | |Baltagi-Li form |92,36 |
|Table 8A, Female baseball games’ attendance, Depending variable is log of attendance, female baseball, regular series 2006 and 2007, n = 220, |
|Standard errors in parenthesis |
|Model: FEMALE |OLS without group dummy variables | | |
| | |REM | |
|LogPrice |0,426 |0,606 | |
| |(0,246) |(0,326) | |
|LogDist |-0,118 |-0,166 | |
| |(0,038)** |(0,034)*** | |
|LogPopH |0,345 |0,387 | |
| |(0,054)*** |(0,108)*** | |
|LogPopV |0,031 |0,037 | |
| |(0,025) |(0,022) | |
|LogPPGH |0,006 |-0,000 | |
| |(0,019) |(0,017) | |
|LogPPGV |0,016 |0,009 | |
| |(0,016) |(0,014) | |
|Temp |0,013 |0,016 | |
| |(0,005)* |(0,005)** | |
|Rain |-0,297 |-0,251 | |
| |(0,074)*** |(0,065) | |
|Roof |0,327 |0,303 | |
| |(0,061)*** |(0,034) | |
|LogUrban |-0,919 |-1,278 | |
| |(0,324)** |(0,599)* | |
|LogPayroll |0,304 |0,416 | |
| |(0,063)*** |(0,103)*** | |
|Saturday |-0,348 |-0,289 | |
| |(0,093)*** |(0,082)*** | |
|Constant |2,608 |2,394 | |
| |(0,961)** |(1,58) | |
|Adjusted R-sq |0,627 | | | | |
|F-test |21,46*** | | | | |
|Diagnostic LL |235,88*** | | | | |
| |Test statistics for the classical model | | |
| |Constant term only |Log likelihood | | | |
| |(1) |= -185,70 | | | |
| |X-variables only (3)|LL = -70,31 | | | |
| |Hypothesis tests | | | | |
| |(3) vs. (1) |230,79*** |(REM) vs. (3) |106,02*** | |
| | | |Baltagi-Li form |92,36 | |
|Table 8B, Female baseball games’ attendance, Depending variable is log of attendance, female baseball, regular series 2006 |
|and 2007, n = 220, Standard errors in parenthesis |
|Model: FEMALE |OLS without group dummy variables |Fixed effects model (FEM) | |
| | | |REM |
|LogPrice |0,374 |0,392 |0,533 |
| |(0,250) |(0,535) |(0,329) |
|LogDist |-0,119 |-0,176 |-0,165 |
| |(0,039)** |(0,034)** |(0,034)*** |
|LogPopH |0,213 |4,730 |0,194 |
| |(0,027)*** |(4,360) |(0,064)** |
|LogPopV |0,032 |0,037 |0,037 |
| |(0,025) |(0,022) |(0,022) |
|LogPPGH |0,009 |0,000 |0,001 |
| |(0,019) |(0,018) |(0,017) |
|LogPPGV |0,016 |0,007 |0,008 |
| |(0,016) |(0,014) |(0,014) |
|Temp |0,013 |0,016 |0,015 |
| |(0,006)* |(0,005)** |(0,005)** |
|Rain |-0,297 |-0,253 |-0,234 |
| |(0,074)*** |(0,065)*** |(0,064)** |
|Roof |0,373 |0,308 |0,343 |
| |(0,061)*** |(0,315) |(0,134)* |
|LogPayroll |0,198 |0,555 |0,303 |
| |(0,052)*** |(0,140)*** |(0,089)*** |
|Saturday |-0,362 |-0,269 |-0,299 |
| |(0,095)*** |(0,082)*** |(0,082)*** |
|Constant |1,102 | |0,0436 |
| |(0,815) | |(1,212) |
|Adjusted R-sq |0,616 | |0,725 | |0,623 |
|F-test |33,05*** | |27,28*** | | |
|Diagnostic LL |235,88*** | | | | |
| |Test statistics for the classical model | | | |
| |Constant term only |Log likelihood | |LM test vs. model (3) |
| |(1) |= -185,70 | |115,63*** |
| |Group effects only |LL = -77,57 | |Hausman test (FEM vs. REM) |
| |(2) | | |16,43 |
| |X-variables only (3)|LL = -74,50 | | | |
| |X- and group effects|LL = -31,93 | | | |
| |(4) | | | | |
| |Hypothesis tests | | | | |
| |(2) vs. (1) |216,27*** | | | |
| |(3) vs. (1) |222,40*** | | | |
| |(4) vs. (1) |307,54*** | | | |
| |(4) vs. (2) |91,28*** | | | |
| |(4) vs. (3) |85,14*** | | | |
|Fixed effects model is not favoured over random effects model |
|Table 8C: Female baseball games’ attendance, Depending variable is log of attendance, female baseball, regular series 2006 |
|and 2007, n = 220, Standard errors in parenthesis |
|Model |SURE |SURE |SURE |
|LogPrice |0,365 |0,365 |0,365 |
| |(0,240) |(0,240) |(0,240) |
|LogDist |-0,110 |-0,110 |-0,110 |
| |(0,041)** |(0,041)** |(0,041)** |
|LogPopH |0,328 |0,328 |0,328 |
| |(0,050)*** |(0,050)*** |(0,050)*** |
|LogPopV |0,023 |0,023 |0,023 |
| |(0,024) |(0,024) |(0,024) |
|LogPPGH |0,009 |0,009 |0,009 |
| |(0,018) |(0,018) |(0,018) |
|LogPPGV |0,018 |0,018 |0,018 |
| |(0,015) |(0,015) |(0,015) |
|Temp |0,013 |0,013 |0,013 |
| |(0,006)* |(0,006)* |(0,006)* |
|Rain |-0,289 |-0,289 |-0,289 |
| |(0,071)*** |(0,071)*** |(0,071)*** |
|Wind |-0,010 |-0,010 |-0,010 |
| |(0,054) |(0,054) |(0,054) |
|Roof |0,327 |0,327 |0,327 |
| |(0,059)*** |(0,059)*** |(0,059)*** |
|LogUrban |-0,795 |-0,795 |-0,795 |
| |(0,317)* |(0,317)* |(0,317)* |
|LogPayroll |0,285 |0,285 |0,285 |
| |(0,062)*** |(0,062)*** |(0,062)*** |
|Monday |0,044 |0,044 |0,044 |
| |(0,198) |(0,198) |(0,198) |
|Tuesday |0,092 |0,092 |0,092 |
| |(0,098) |(0,098) |(0,098) |
|Wednesday |-0,008 |-0,008 |-0,008 |
| |(0,059) |(0,059) |(0,059) |
|Thursday |0,122 |0,122 |0,122 |
| |(0,088) |(0,088) |(0,088) |
|Friday |0,183 |0,183 |0,183 |
| |(0,127) |(0,127) |(0,127) |
|Saturday |-0,330 |-0,330 |-0,330 |
| |(0,094)*** |(0,094)*** |(0,094)*** |
|Constant |2,57 |2,57 |2,57 |
| |(0,945)** |(0,945)** |(0,945)** |
|Depending variable is log of attendance, female baseball, regular series 2006 and 2007, n = 220 |
|Standard errors in parenthesis |
|Adjusted R-sq |0,627 | | |
|F-test |21,46*** | | | | | |
|Diagnostic LL |255,75*** | | | | | |
| | | | | | |
| | | | | | | |
|LogPopH |0,029 | | | |0,013 | |
| |(0,007)*** | | | |(0,013) | |
|LogPayroll |0,009 | |-0,011 | | | |
| |(0,013) | |(0,015) | | | |
|LogUrban | | |0,186 | |0,109 |0,170 |
| | | |(0,042)*** | |(0,070) |(0,037)*** |
|Roof |0,030 | |0,041 | |0,034 |0,039 |
| |(0,016) | |(0,015)** | |(0,016)* |(0,015)** |
|constant |1,52 | |1,20 | |1,29 |1,16 |
| |(0,138)*** | |(0,171)*** | |(0,207) |(0,160)*** |
|Depending variable is LogPrice |
|Table 9A: SURE results, Female baseball, standard errors in parenthesis |
|Model |SURE |SURE |
|LogPrice |-5,33 |-5,88 |
| |(0,653)*** |(0,24)*** |
|LogDist |-0,125 |-0,131 |
| |(0,240)*** |(0,037)*** |
|LogPopH |0,376 |0,387 |
| |(0,037)*** |(0,039)*** |
|LogPopV |0,031 | |
| |(0,024) | |
|LogPPGH |0,016 | |
| |(0,018) | |
|LogPPGV |0,015 | |
| |(0,015) | |
|Temp |0,012 |0,016 |
| |(0,005)* |(0,005)** |
|Rain |-0,318 |-0,300 |
| |(0,071)*** |(0,070)*** |
|Wind | | |
|Roof |0,611 |0,657 |
| |(0,105)*** |(0,111)*** |
|LogUrban | | |
|LogPayroll | | |
|Monday | | |
|Tuesday | | |
|Wednesday | | |
|Thursday | | |
|Friday | | |
|Saturday |-0,365 |-0,348 |
| |(0,091)*** |(0,091)*** |
|Constant |12,5 |13,7 |
| |0,653)*** |(0,603)*** |
|Adjusted R-sq | | |
|F-test | | | | |
|Diagnostic LL |-213,16 | |-230,10 | |
|Depending variable is log of attendance, female baseball, regular series 2006 and 2007, n = 220, |
|Standard errors in parenthesis |
| | | | | |
|LogPopH | | | | |
|LogPayroll |-0,042 | |-0,043 | |
| |(0,008)*** | |(0,008)*** | |
|LogUrban |0,189 | |0,184 | |
| |(0,033)*** | |(0,032)*** | |
|Roof |0,050 | |0,050 | |
| |(0,015)*** | |(0,015)** | |
|constant |1,52 | |1,55 | |
| |(0,125)*** | |(0,120)*** | |
|depending variable is LogPrice |
|Table 9B, SURE results, Female baseball, standard errors in parenthesis |
|Variable |
|Variable |
|Model |OLS without group dummy variables |Random effects model (REM) | | |
| | | |OLS |REM |
|LogPrice |1,43 |0,287 |1,42 |0,450 |
| |(0,279)*** |(0,468) |(0,273)*** |(0,479) |
|LogDist |-0,150 |-0,165 |-0,162 |-0,176 |
| |(0,030)*** |(0,028)*** |(0,029)*** |(0,027)*** |
|LogPopH |0,015 |0,023 |0,023 |0,031 |
| |(0,032) |(0,068) |(0,032) |(0,072) |
|LogPopV |-0,050 |-0,048 |-0,051 |-0,050 |
| |(0,020)* |(0,018)** |(0,020)* |(0,018)* |
|LogPPGH |0,015 |0,034 |0,017 |0,035 |
| |(0,020) |(0,020) |(0,020) |(0,020) |
|LogPPGV |0,031 |0,018 |0,030 |0,018 |
| |(0,022) |(0,021) |(0,021) |(0,020) |
|Temp |0,009 |0,009 |0,008 |0,009 |
| |(0,005) |(0,004)* |(0,006) |(0,004)* |
|Rain |-0,185 |-0,183 |-0,169 |-0,174 |
| |(0,068)** |(0,063)** |(0,067)* |(0,062)* |
|Wind |0,074 |0,012 | | |
| |(0,046) |(0,045) | | |
|Roof |0,070 |0,072 |0,061 |0,072 |
| |(0,043) |(0,086) |(0,042) |(0,091) |
|LogUrban |-0,729 |-0,644 |-0,759 |-0,667 |
| |(0,183)*** |(0,374) |(0,182)*** |(0,393) |
|LogPayroll |-0,276 |0,162 |-0,280 |0,220 |
| |(0,141) |(0,221) |(0,140)* |(0,223) |
|Tuesday |0,068 |0,108 | | |
| |(0,260) |(0,242) | | |
|Wednesday |-0,207 |-0,065 | | |
| |(0,307) |(0,286) | | |
|Thursday |0,034 |0,082 | | |
| |(0,260) |(0,242) | | |
|Friday |-0,077 |0,017 | | |
| |(0,270) |(0,252) | | |
|Saturday |-0,233 |-0,201 | | |
| |(0,302) |(0,279) | | |
|Sunday |0,069 |0,102 | | |
| |(0,259) |(0,241) | | |
|Constant |11,466 |7,639 |11,765 |7,433 |
| |(1,81)*** |(3,00)* |(1,77)*** |(3,07)* |
|Depending variable is log of attendance, male baseball, regular series 2006 and 2007, n = 325 |
|Standard errors in parenthesis |
|Adjusted R-sq |0,280 | |0,270 |
|F-test |8,01*** | | | |11,87 | |
|Diagnostic LL |-117,04 | | | |-123,11 | |
| |Test statistics for the classical model | | |Test statistics for the classical model |
| |Constant term only |Log likelihood | | |Constant term only |Log likelihood |
| |(1) |= -179,77 | | |(1) |= -179,77 |
| |X-variables only (3)|LL = -117,04 | | |X-variables only (3)|LL = -123,11 |
| |Hypothesis tests | | | |Hypothesis tests | |
| |(3) vs. (1) |125,46*** |(REM) vs. (3) |23,30*** |(3) vs. (1) |113,33*** |
| | | |Baltagi-Li form |19,75 | | |
| | | | | |(REM) vs. (3) |43,88*** |
| | | | | |Baltagi-Li form |37,21 |
|Table 11A, Male baseball, regular series 2007 and 2008, N = 325, standard errors in parenthesis |
|Model: MALE |OLS without group dummy variables |Fixed effects model (FEM) | |
| | | |REM |
|LogPrice |1,33 |-1,39 |0,454 |
| |(0,281)*** |(0,909) |(0,501) |
|LogDist |-0,144 |-0,175 |-0,165 |
| |(0,030)*** |(0,028)*** |(0,028)*** |
|LogPopH |-0,082 |0,015 |-0,064 |
| |(0,021)*** |(1,42) |(0,049) |
|LogPopV |-0,051 |-0,050 |-0,049 |
| |(0,020)* |(0,018)** |(0,018)** |
|LogPPGH |0,024 |0,042 |0,037 |
| |(0,021) |(0,020)* |(0,020) |
|LogPPGV |0,027 |0,015 |0,016 |
| |(0,022) |(0,021) |(0,021) |
|Temp |0,008 |0,009 |0,009 |
| |(0,005) |(0,004) |(0,004)* |
|Rain |-0,202 |-0,177 |-0,186 |
| |(0,068)** |(0,064)** |(0,063)** |
|Vind |0,073 |-0,001 |0,009 |
| |(0,047) |(0,046) |(0,045) |
|LogPayroll |-0,070 |0,669 |0,338 |
| |(0,134) |(0,347) |(0,225) |
|Tuesday |0,170 |0,057 |0,122 |
| |(0,265) |(0,244) |(0,241) |
|Wednesday |-0,148 |-0,052 |-0,043 |
| |(0,314) |(0,287) |(0,285) |
|Thursday |0,126 |0,034 |0,095 |
| |(0,265) |(0,244) |(0,242) |
|Friday |0,054 |0,027 |0,047 |
| |(0,274) |(0,253) |(0,251) |
|Saturday |-0,204 |-0,234 |-0,201 |
| |(0,309) |(0,279) |(0,278) |
|Sunday |0,161 |0,049 |0,115 |
| |(0,263) |(0,243) |(0,240) |
|Constant |6,94 | |3,89 |
| |(1,40)*** | |(2,44) |
|Depending variable is log of attendance, male baseball, regular series, 2006 and 2007, n = 325 |
|Adjusted R-sq |0,246 | |0,394 | |0,248 |
|F-test |7,60*** | |8,27*** | | |
|Diagnostic LL |-125,70*** | | | | |
| |Test statistics for the classical model | | | |
| |Constant term only |Log likelihood | |LM test vs. model (3) |
| |(1) |= -179,77 | |63,15*** |
| |Group effects only |LL = -128,71 | |Baltagi-Li form of LM statistics = 52,70 |
| |(2) | | | |
| |X-variables only (3)|LL = -125,70 | | | |
| |X- and group effects|LL = -83,05 | | | |
| |(4) | | | | |
| |Hypothesis tests | | | | |
| |(2) vs. (1) |102,13*** | | | |
| |(3) vs. (1) |108,14*** | | | |
| |(4) vs. (1) |194,44*** | | | |
| |(4) vs. (2) |91,31** | | | |
| |(4) vs. (3) |85,30*** | | | |
|Table 11B, Male baseball, regular series 2007 and 2008, N = 325, standard errors in parenthesis |
|Model: MALE |OLS without group dummy variables |Fixed effects model (FEM) | |
| | | |REM |
|LogPrice |1,25 |-1,48 |0,287 |
| |(0,276)*** |(0,884) |(0,506) |
|LogDist |-0,153 |-0,178 |-0,171 |
| |(0,030)*** |(0,027)*** |(0,027)*** |
|LogPopH |-0,079 |-0,089 |-0,059 |
| |(0,021)*** |(1,39) |(0,051) |
|LogPopV |-0,049 |-0,051 |-0,049 |
| |(0,020)* |(0,018)** |(0,018)** |
|LogPPGH |0,027 |0,043 |0,039 |
| |(0,020) |(0,020)* |(0,019)* |
|LogPPGV |0,022 |0,013 |0,015 |
| |(0,022) |(0,020) |(0,020) |
|Temp |0,007 |0,009 |0,009 |
| |(0,005) |(0,004)* |(0,004)* |
|Rain |-0,188 |-0,174 |-0,182 |
| |(0,067)** |(0,063)** |(0,062)** |
|Vind | | | |
|LogPayroll |-0,046 |0,700 |0,407 |
| |(0,133) |(0,334)* |(0,225) |
|Saturday |-0,349 |-0,272 |-0,296 |
| |(0,0167)* |(0,154) |(0,152) |
|Constant |6,98 | |3,52 |
| |(1,37)*** | |(2,47) |
|Depending variable is log of attendance, male baseball, regular series, 2006 and 2007, n = 325 |
|Adjusted R-sq |0,240 | |0,405 | |0,226 |
|F-test |11,26** | | | | |
|Diagnostic LL | | | | | |
| |Test statistics for the classical model | | | |
| |Constant term only |Log likelihood | |LM test vs. model (3) |
| |(1) |= -179,77 | |92,00*** |
| |Group effects only |LL = -128,71 | |Hausman test (FEM vs. REM) |
| |(2) | | |14,81 not sig. (prob = 0,139) |
| |X-variables only (3)|LL = -129,96 | | | |
| |X- and group effects|LL = -83,44 | | | |
| |(4) | | | | |
| |Hypothesis tests | | | | |
| |(2) vs. (1) |102,13*** | | | |
| |(3) vs. (1) |99,63*** | | | |
| |(4) vs. (1) |192,67*** | | | |
| |(4) vs. (2) |90,54*** | | | |
| |(4) vs. (3) |93,04*** | | | |
|Table 11C, Male baseball, regular series 2007 and 2008, N = 325, standard errors in parenthesis |
|Model: MALE |OLS without group dummy variables |Fixed effects model (FEM) | |
| | | |REM |
|LogPrice |1,04 |-1,52 |0,098 |
| |(0,275)*** |(0,856) |(0,523) |
|LogDist |-0,159 |-0,178 |-0,172 |
| |(0,031)*** |(0,027)*** |(0,027)*** |
|LogPopH | | | |
|LogPopV |-0,045 |-0,052 |-0,050 |
| |(0,020)* |(0,018)** |(0,018)** |
|LogPPGH |0,046 |0,051 |0,048 |
| |(0,017)** |(0,016)** |(0,016)** |
|LogPPGV | | | |
|Temp |0,007 |0,009 |0,009 |
| |(0,005) |(0,004)* |(0,004)* |
|Rain |-0,188 |-0,172 |-0,180 |
| |(0,069)** |(0,062)** |(0,062)** |
|Vind | | | |
|LogPayroll |0,065 |0,712 |0,504 |
| |(0,132) |(0,332)* |(0,227)* |
|Saturday |-0,355 |-0,275 |-0,289 |
| |(0,171)* |(0,154) |(0,152)* |
|Constant |5,31 | |2,20 |
| |(1,32) | |(2,36) |
|Depending variable is log of attendance, male baseball, regular series, 2006 and 2007, n = 325 |
|Adjusted R-sq |0,211 | |0,408 | |0,194 |
|F-test |11,86** | |11,63** | | |
|Diagnostic LL | | | | | |
| |Test statistics for the classical model | | | |
| |Constant term only |Log likelihood | |LM test vs. model (3) |
| |(1) |= -179,77 | |146,75 |
| |Group effects only |LL = -128,71 | |Hausman test (FEM vs. REM) |
| |(2) | | |9,85 (not sign. prob = 0,275) |
| |X-variables only (3)|LL = -137,10 | | | |
| |X- and group effects|LL = -83,69 | | | |
| |(4) | | | | |
| |Hypothesis tests | | | | |
| |(2) vs. (1) |102,13*** | | | |
| |(3) vs. (1) |85,34*** | | | |
| |(4) vs. (1) |192,17*** | | | |
| |(4) vs. (2) |90,04*** | | | |
| |(4) vs. (3) |106,83*** | | | |
|Table 11D: Male baseball, regular series 2007 and 2008, N = 325, standard errors in parenthesis |
|Model |SURE |SURE | | | | |
|LogPrice |1,43 |1,05 |1,36 |1,36 |2,39 |1,28 |
| |(0,271)*** |(0,208)*** |(0,268)*** |(0,268)*** |(0,267)*** |(0,275)*** |
|LogDist |-0,150 |-0,153 |-0,159 |-0,159 |-0,158 |-0,153 |
| |(0,029)*** |(0,029)*** |(0,029)*** |(0,029)*** |(0,029)*** |(0,029) |
|LogPopH |0,015 |-0,076 |0,020 |0,020 | |-0,079 |
| |(0,031) |(0,020)*** |(0,032) |(0,032) | |(0,021)*** |
|LogPopV |-0,050 |-0,049 |-0,049 |-0,049 |-0,049 |-0,049 |
| |(0,019)** |(0,020)* |(0,019)* |(0,019)* |(0,019)* |(0,020)* |
|LogPPGH |0,015 |0,028 |0,018 |0,018 |0,017 |0,027 |
| |(0,020) |(0,020) |(0,020) |(0,020) |(0,020) |(0,020) |
|LogPPGV |0,031 |0,021 |0,029 |0,029 |0,029 |0,021 |
| |(0,021) |(0,021) |(0,021) |(0,021) |(0,021) |(0,021) |
|Temp |0,009 |0,007 |0,008 |0,008 |0,008 |0,008 |
| |(0,004)* |(0,005) |(0,004) |(0,004) |(0,004) |(0,004) |
|Rain |-0,185 |-0,180 |-0,172 |-0,172 |-0,172 |-0,179 |
| |(0,066)** |(0,067)** |(0,066)** |(0,066)** |(0,066)** |(0,068)** |
|Wind |0,074 | | | | | |
| |(0,044) | | | | | |
|Roof |0,070 |0,022 |0,057 |0,057 |0,080 |0,029 |
| |(0,042) |(0,042) |(0,041) |(0,041) |(0,042) |(0,042) |
|LogUrban |-0,729 | |-0,729 |-0,729 |-0,715 | |
| |(0,178)*** | |(0,179)*** |(0,179)*** |(0,116)*** | |
|LogPayroll |-0,276 | |-0,257 |-0,257 |-0,578 |-0,045 |
| |(0,137)* | |(0,138) |(0,138) |(0,138)*** |(0,131) |
|Tuesday |0,068 | | | | | |
| |(0,253) | | | | | |
|Wednesday |-0,207 | | | | | |
| |(0,298) | | | | | |
|Thursday |0,034 | | | | | |
| |(0,252) | | | | | |
|Friday |-0,077 | | | | | |
| |(0,262) | | | | | |
|Saturday |-0,233 |-0,339 |-0,282 |-0,282 |-0,283 |-0,344 |
| |(0,293) |(0,164)* |(0,161) |(0,161) |(0,161) |(0,165)* |
|Sunday |0,069 | | | | | |
| |(0,251) | | | | | |
|Constant |11,47 |6,84 |11,49 |11,49 |13,20 |6,89 |
| |(1,75)*** |(0,542)*** |(1,74)*** |(1,74)*** |(1,66)*** |(1,35*** |
|Depending variable is log of attendance, male baseball, regular series 2006 and 2007, n = 325 |
|Standard errors in parenthesis |
|Adjusted R-sq |0,280 |0,240 |0,274 |0,274 |0,242 |0,239 |
|F-test |8,01*** |11,21*** |11,19*** |11,19*** |10,40*** |10,26** |
|Diagnostic LL |-107,25*** |-124,55*** |-114,95*** |-114,95*** |-123,03*** |-123,61** |
| |
|LogPopH |0,011 |0,010 |0,011 |0,016 |0,016 |0,016 |
| |(0,006) |(0,007) |(0,007) |(0,004)*** |(0,004)*** |(0,004)*** |
|LogPayroll |0,323 |0,325 |0,323 |0,313 |0,313 |0,313 |
| |(0,022)*** |(0,022)*** |(0,022)*** |(0,020)*** |(0,020)*** |(0,020)*** |
|LogUrban |0,034 |0,043 |0,034 | | | |
| |(0,037) |(0,037) |(0,037) | | | |
|Roof |-0,025 |-0,025 |-0,025 |-0,023 |-0,023 |-0,023 |
| |(0,008)** |(0,008)** |(0,008)** |(0,008)*** |(0,008)** |(0,008)** |
|constant |-1,93 |-1,98 |-1,93 |-1,72 |-1,72 |-1,72 |
| |(0,339)*** |(0,339)*** |(0,339)*** |(0,247)*** |(0,246)*** |(0,247)*** |
|depending variable is LogPrice | |
|Adjusted R-sq |0,458 |0,458 |0,458 |0,458 |0,458 |0,458 |
|F-test |64,49*** |69,47*** |69,49*** |92,44*** |92,44*** |92,44*** |
|Diagnostic LL |386,50*** |386,47*** |386,50*** |385,58*** |385,58*** |358,58*** |
|Table 12: SURE estimations, male baseball |
| |Daily |Several times a week|Several times a |Occasionally |Never |Total |
| | | |month | | | |
|How often do you |4 (0,3%) |17 (1,3%) |82 (6,2%) |691 (52,3%) |526 (39,8%) |1320 |
|attend a sports | | | | | | |
|activity? | | | | | | |
|How often do you |0 |4 (1,4%) |20 (7,2%) |152 (54,5%) |103 (36,9%) |279 |
|attend a sports |(F: 0 – M:0) |(F: 0 – M:4) |(F: 9 – M:11) |(F: 62 – M: 90) |(F: 78 – M : 25) |(F:149 – M: 130) |
|activity? (Single)* | | | | | | |
|(21,1%) | | | | | | |
|How often do you |3 (0,5%) |9 (1,4%) |39 (6,1%) |343 (53,8%) |243 (38,1%) |637 |
|attend a sports |(F: 0 – M: 3) |(F: 3 – M: 6) |(F: 17 – M: 21) |(F: 162 – M: 181) |(F: 170 – M: 73) |(F: 352 – M: 284) |
|activity? (Married | | | | | | |
|or registered | | | | | | |
|couple)** | | | | | | |
|(48,3%) | | | | | | |
|How often do you |0 |3 (1,3%) |21 (9,1%) |126 (54,6%) |81 (35,1%) |231 |
|attend a sports |(F: 0 – M:0) |(F:2 – M: 1) |(F: 11 – M: 10) |(F: 60 – M: 66) |(F: 50 – M: 30) |(F:123 – M: 107) |
|activity? | | | | | | |
|(cohabitation | | | | | | |
|without marriage)***| | | | | | |
|(17,5%) | | | | | | |
|How often do you |0 |0 |0 |3 (50%) |3 (50%) |6 |
|attend a sports |(F: 0 – M: 0) |(F: 0 – M: 0) |(F: 0 – M: 0) |(F: 1 – M: 2) |(F: 1- M: 2) |(F: 2 – M:4) |
|activity? | | | | | | |
|(Separated) | | | | | | |
|(0,5%) | | | | | | |
|How often do you |0 |1 (0,9%) |2 (1,9%) |48 (45,3%) |55 (51,2%) |106 |
|attend a sports |(F: 0 – M: 0) |(F: 0 – M: 1) |(F: 1 – M:1) |(F: 33 – M: 15) |(F: 45 – M: 10) |(F: 79 – M: 27) |
|activity? | | | | | | |
|(divorced)**** | | | | | | |
|(8,0%) | | | | | | |
|How often do you |1 (2,2%) |0 |0 |13 (28,3%) |32 (69,6%) |46 |
|attend a sports |(F: 1 – M: 0) |(F: 0 – M: 0) |(F: 0 – M: 0) |(F: 9 – M: 4) |(F: 25 – M: 7) |(F: 35 – M: 11) |
|activity? | | | | | | |
|(Widowed)***** | | | | | | |
|(3,5%) | | | | | | |
|Appendix 1: Sports consumption in Finland 2007, International Social Survey Programme, own calculations |
|*Significantly different between genders, Mann-Whitley U-test, (z= -5,507, sig. = 0,000) |
|**Significantly different between genders, Mann-Whitney U-test (z=-5,853, sig.=0,000) |
|*** Not different between genders, Mann-Whitney U-test (z = -1,621, sig. = 0,105) |
|**** Significantly different between genders, Mann-Whitney U-test (z=-1,974, sig.=0,048) |
|***** Not different between genders, Mann-Whitney U-test (z = -0,418, sig. = 0,676) |
|20 Biggest cities Jan2007 |
-----------------------
[1] Finnish baseball is different than baseball played in USA. For the rules, see
[2] In Finland the sample has 1354 valid answers, data collected through mail questionnaire during 18.9. – 11.12.2007
[3] Recent (February – March 2007) Eurobarometer 67.1 reports that almost 56 % in the sample (n = 1054 in Finland) had not attended any sport event during the last 12 month period. The figure was lower for male (44%) than for female (65%).
[4] χ2 = 13,78 (sig. 0,000)
[5] A good survey, see Borland & Macdonald 2003 or Simmons 2006
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.