Gender differences in cultural and sports activities attendance: an intra-couple analysis

ABSTRACT

We study how education affects cultural and sports attendance of spouses using the Survey on Living Conditions 2015. With a sample of 7920 heterosexual couples, we estimate bivariate models for attendance to cinema, visits to monuments, life performances and sports events. We conduct research at the intra-household level and estimate the marginal effect of female and male characteristics over the probability of having a household where both members participate in the same type of social activity. Own and spouse’s education, income and having young children (0–3 years old) are the main drivers of both partners’ attending the same activities. Both male and female education have a positive, but not statistically different effect, on these probabilities. The probability of both spouses participating in culture and sports activities increases with income and decreases with the presence of young children.

KEYWORDS:

• Cultural participation
• gender
• intra-couple decisions
• leisure participation
• sports attendance

Notes

¹ Each individual, aged 16 or over on December 31st of the year prior to the survey, is asked if during the last 12 months he/she went to the cinema, live performances, museums or sports. Based on that information, we can identify if both partners attended at the same type of activity or not, though we cannot identify whom he/she has gone with and if the members of a couple attended together at all or at some of the leisure event.

² In order to consider all the individuals in our sample, we include an additional dummy variable to control for individuals that do not report a current or last occupation.

³ When compared to the traditional and more explored source of information on cultural practices in Spain, the Survey on Cultural Habits and Practices (SCHP), this dataset is much more suitable for several reasons. First, it provides information about all members in the household above 16 years of age. Second, a wide set of variables that measure the income of the members of the household is available. Third, the complete dataset includes information regarding around 32,000 individuals living in 13,000 households.

⁴ It includes historical monuments, museums, art galleries, archaeological sites, etc. if the purpose of the visit was to become acquainted with the cultural content of the site.

⁵ Live performances include theatre plays, concerts, operas, ballet, dance performances, etc., whether performed by professionals or amateurs. Going to live performances to watch one’s own children is included and street performances are also included if they are organized events. Sport events include any organized sport event performed by professionals or amateurs as well as attending an event to watch one’s own children.

⁶ As mentioned by Mauri and Wolf (2020) ‘data on actual joint attendance are rare and “both attended” does not automatically imply joint attendance’.

⁷ Table of descriptive statistics available upon request.

⁸ This scale assigns a value of 1 to the household head, 0.5 to each additional household member older than 13, and 0.3 to each child younger than 14, as proposed by Hagenaars, De Vos, and Zaidi (1994).

⁹ The estimated coefficients of the bivariate probit models, one for each leisure activity, are available upon request.

Additional information

Funding

This work was supported by the Basque Government [IT336-19] and the Spanish Ministry of Science and Innovation [PID2019-108718GB-I00]