Archers predicting their score: Isik 2018 study

One hundred and sixteen archers competing at the Turkish adult and youth indoor championship in Izmir completed a questionnaire on motivational orientation before shooting and wrote down the score they expected, along with their bib number, so that after the event the prediction could be compared with the result. A cluster analysis split them into two groups: 34 (29.3%) with "low" orientation and 82 (70.7%) with "high" orientation.

The high orientation group finished with a mean of 560.15 points and the low orientation group with 431.8 (t = 2.849, p < 0.05). In the high group the predicted score is associated with the actual score (R² = 0.305, F = 4.173, p < 0.05; each predicted point is worth 0.986 actual points on average), while in the low group it is not (R² = 0.160, F = 0.825, p > 0.05). From this the author concludes that those with higher orientation perform better and predict their own result more accurately.

The participants were 82 men and 34 women who had qualified for the national championship through regional events; the reported mean age is 33.29. Before shooting they completed the Turkish version of the TEOSQ, a 13-item questionnaire with 7 items on task orientation (feeling successful when improving) and 6 on ego orientation (feeling successful when beating others), answered from 1 to 5. A form designed by the author collected personal details, bib number and predicted score; anyone unwilling to give their number or prediction was excluded.

To form the groups the author used hierarchical cluster analysis with Ward's method followed by k-means with two clusters; differences between the groups were checked with analysis of variance, and the link between predicted and actual score with a separate linear regression for each group. The divisions (recurve, compound, barebow), age classes, distance, number of arrows and maximum possible score are not reported.

The TEOSQ measures two separate orientations, task and ego, which can both be high or both be low in the same person. The study merges them into a single "high" or "low" orientation that the questionnaire does not provide for, and the author acknowledges not having looked at the two separately. The comparison between prediction and result, announced as the subject of the study, is never actually computed: a regression of actual score on predicted score says that those who expect more tend to score more, not how close the prediction is to the result.

Many figures do not add up. The item means in the two clusters reach 56.3 in one group and stay below 1 in the other, values that are impossible on a 1 to 5 scale, and the group the author calls "low", the 34, is the one with the higher means; in the text the participants become 166; a mean age of 33.29 with a standard deviation of 1.26 is not credible at an adult and youth championship; and with 82 people an R² of 0.305 would give an F far larger than 4.173. Above all, recurve, compound and barebow archers, juniors and adults score very differently: the gap between 560 and 432 points could come entirely from how the two groups are made up. The journal is flagged for questionable editorial standards.

For an archer the study offers nothing usable. That better archers also predict higher scores is close to obvious, and it says nothing about whether writing down a target before a competition, or training a particular motivational orientation, is useful. If the idea is of interest, the useful figure would be the gap between prediction and result, archer by archer and within the same division, and that is exactly what is missing here.