Stroemer, Ellen (2024) Can Soccer Scouts Identify Talent Better Than Laypeople? Master thesis, Psychology.
|
Text
Master-Thesis-s4228790-Can-Scouts-Predict-Talent-Better-Than-Laypeople.pdf Download (1MB) | Preview |
Abstract
In this study, the validity of the performance predictions of eighteen soccer scouts and 50 laypeople in soccer was assessed and compared. The application of the lens model as a theoretical framework allowed for various insights into the decision-making process of the judges. The judgment task entailed 50 real player profiles that consisted of four informational cues (e.g., tackling, interceptions, sprinting speed and game insight). The judges were asked to consider these scores to make predictions about the player’s worth. The actual market value of the players served as the criterion that the predictions were compared to. The analysis showed that the scouts did not make more accurate predictions than the laypeople, which provides evidence against the expert’s eye for talent. Furthermore, nearly all judges were outperformed by their own model, which administered the average weights that the judges assigned to the four cues. This finding confirmed that intuitive judgments are impacted by noise. That implies that the judges tend to weigh the cues inconsistently across cases, which reduces their validity. Therefore, applying a decision rule to combine information seems to be beneficial as it removes this noise. Moreover, the scouts were not able to capture valid insights that a model approach could not, conveying that the employment of a decision rule does not harm the validity of the judgment. Lastly, recommendations on effective and autonomy-preserving approaches to talent assessment are provided.
| Item Type: | Thesis (Master) |
|---|---|
| Supervisor name: | Meijer, R.R. and Peringa, I.P. |
| Degree programme: | Psychology |
| Differentiation route: | Talent Development and Creativity (TDC) [Master Psychology] |
| Date Deposited: | 12 Jul 2024 09:53 |
| Last Modified: | 12 Jul 2024 09:53 |
| URI: | http://gmwpublic.studenttheses.ub.rug.nl/id/eprint/3853 |
Actions (login required)
 |
View Item |