Media Briefings

CHILDREN DO NOT BEHAVE LIKE ADULTS: New evidence on gender gaps in performance and risk-taking from TV game shows

  • Published Date: October 2016

While pre-teen girls and boys are equally willing to take risks in high-stakes situations, adult women have less of an appetite for risk than men. But if girls find themselves in a male-dominated environment, they are likely both to take on more risk and to perform worse than they would do in an all-female or mixed gender environment.

These are among the findings of research by Jenny Säve-Söderbergh and Gabriella Sjögren Lindquist, which is forthcoming in the Economic Journal. Their study uses data from the TV game-show Jeopardy and the equivalent show for children Junior Jeopardy to explore differences in performance and risk-taking between children and adults and between males and females.

This study provides unique field evidence from a high-stakes environment of female behaviour being affected by gender context as early as 10-11 years of age. If girls are randomly assigned to an all-boys environment, they perform cognitively worse and take on more economic risk, compared with an environment in which there are other girls.

Analysing additional data on adult behaviours in the same field setting, the researchers provide new evidence of there being no gender gap in willingness to take risk among girls and boys, whereas there is a gap between adult males and females (as previous studies have shown). This indicates either that gender gaps in risk-taking evolve with age or that they differ by cohort.

These results are important as both cognitive performance and risk preferences are bound to shape many important decisions, such as educational choices and labour market outcomes. It is thus crucial to know if there are gender gaps in these behaviours; if so, when these gaps arise; and importantly, if these behaviours are susceptible to the environment within which they are executed.

To study children’s and adults’ cognitive performance and risk preferences, the study makes use of a high-stakes field setting – the Daily Double – arising in the Swedish version of Jeopardy and Junior Jeopardy.

Winners in these shows earn stakes equivalent to a half-month’s salary. In the random event of a Daily Double, a contestant is individually faced with a decision to wager his or her score on the probability of answering an ensuing question correctly. By evaluating these Daily Doubles, it is possible to elicit both risk preferences and cognitive performance through wagers and correct answers; and, what’s vital, in a situation in which the behaviour of co-contestants is irrelevant.

In addition, to identify the causal effect of gender context on wagering and performance, the researchers make use of the fact that contestants are randomly assigned to co-contestants, thus creating a natural experiment of decision-making in different gender contexts.

Despite the high-stakes setting and the lack of strategic advantage, the study documents that girls perform worse and employ greater but less gainful risk-taking when they are randomly assigned only to boy opponents, compared with when they are randomly assigned to an only-girls or mixed-gender group of opponents. Conversely, adult women wager less but do not perform differently if randomly assigned to a male-dominated context.

The researchers find no robust effects of gender context on either men’s or boys’ risk-taking or performance. Furthermore, while girls and boys are equally willing to take risks in their wagering, women take less risk than men. Notably, the largest within-gender difference in risk-taking is found among males, with boys taking the lowest level of risk of all contestants and men wagering the most.

This study provides new field evidence on gender gaps in risk-taking and on whether male- and female-dominated contexts matter for the behaviour of girls and women. It also contributes unique evidence on children’s economic decision-making from a high-stakes setting. In addition, the researchers analyse gender gaps with data that were neither designed nor gathered to analyse gender per se.

ENDS


Notes for editors: ‘Children Do Not Behave Like Adults: Gender Gaps in Performance and Risk Taking in a Random Social Context in the High-Stakes Game Shows Jeopardy and Junior Jeopardy’ by Jenny Säve-Söderbergh and Gabriella Sjögren Lindquist is forthcoming in the Economic Journal.

Jenny Säve-Söderbergh and Gabriella Sjögren Lindquist are at the Swedish Institute for Social Research at Stockholm University.

For further information: contact Jenny Säve-Söderbergh on +46761411312 (email: Jenny.Save-Soderbergh@sofi.su.se); or Romesh Vaitilingam on +44-7768-661095 (email: romesh@vaitilingam.com; Twitter: @econromesh).