3.1. Descriptive Statistics and Practice
We open the results section by providing descriptive statistics and comparisons between the male and female respondents in our sample.
Table 1 shows that the male keepers in our sample are significantly older than the female keepers (t = 2.75,
p = 0.007, df = 93). The male keepers also had more years of playing experience in general (t = 2.36,
p = 0.020, df = 93) and at a professional level in particular (t = 2.85,
p = 0.005, df = 93). No significant differences were observed between the groups in terms of the number of respondents who participated in international matches (χ
2 = 0.62,
p = 0.431, df = 1), in EHF Champions League (χ
2 = 0.69,
p = 0.403, df = 1), played for a national team (χ
2 = 0.00,
p = 0.964, df = 1), took part in European or other continental championship (χ
2 = 1.96,
p = 0.161, df = 1), at the World Handball Championship (χ
2 = 0.44,
p = 0.504, df = 1).
Table 2 presents the number of weekly training sessions in which empty-goal situations were practiced.
We used a chi-square test to verify that there were no significant or near-significant differences between men and women regarding the amount of weekly practice; the obtained result was (χ
2 = 3.49,
p = 0.479, df = 4), (χ
2 = 2.73,
p = 0.603, df = 4), and (χ
2 = 3.38,
p = 0.496, df = 4) for the first, second, and third practice questions, respectively.
Table 2 shows that few goalkeepers practiced an empty-goal situation more than twice a week. For further analysis, we used the average for the three practice questions and reduced the number of categories down to three (
Table 3): no practice (average = 0), n = 19 (20.0%) goalkeepers; once a week (average between 0 to 1.99), n = 50 (52.6%) goalkeepers; twice or more per week (average larger than 2), n = 26 (27.4%) goalkeepers. Once again, we conducted a chi-square test to identify gender-based differences in the new practice variable (χ
2 = 0.12,
p = 0.939, df = 2). Additionally, we tested whether participation in a national team is related to the number of weekly practices of empty-goal scenarios and found no significant correlation (χ
2 = 2.03,
p = 0.362, df = 2).
3.2. Confidence, Anxiety, Sense of Belonging, and Overall Opinion
We continue by presenting the degree of confidence exhibited by male and female goalkeepers regarding various abilities that are essential while playing empty goal.
A two-way ANOVA was conducted to evaluate the effect of gender and practice on the average of the 10 confidence items that are presented in
Table 3. Male keepers (7.41) seem to be more confident than female keepers (6.97); however, this difference failed to reach significance (f = 1.88,
p = 0.173, df = 1). The amount of practice was not found to significantly contribute to the keepers’ confidence level (f = 0.18,
p = 0.832, df = 2), and there was no interaction between gender and practice (f = 0.134,
p = 0.875, df = 2).
A two-way MANOVA was used to explore the effect of gender, practice, and interaction between these two factors on the confidence of the respondents on each of the 10 items (see
Appendix A). No significant differences were observed between male and female keepers across the 10 confidence questions (f = 0.53,
p = 0.864, df = 10, 80). The practice factor was found to be near-significant (f = 1.56,
p = 0.067, df = 20, 160). We can see in
Appendix A that keepers (both women and men) who practice empty-goal scenarios at least twice a week exhibited a higher level of confidence regarding their ability to maintain momentum despite frequent empty-goal substitutions (f = 4.23,
p = 0.018, df = 2). No significant interaction between gender and practice was detected (f = 1.42,
p = 0.120, df = 20, 160). One item, however, was shown to have significant interaction; we found that male keepers’ confidence regarding their ability to save a long-distance shot was positively correlated with the number of weekly practices of empty-goal scenarios, whereas female keepers exhibited a negative correlation (f = 3.86,
p = 0.025, df = 2).
The effect of participation in a national team on the respondents’ confidence was evaluated through a separate two-way MANOVA. Participation in a national team was shown to be a non-significant factor (f = 0.73, p = 0.693, df = 10, 82), gender was shown to be non-significant (f = 0.74, p = 0.676, df = 10, 82), and the interaction between gender and national team participation was shown to be non-significant (f = 0.73, p = 0.689, df = 10, 82).
The goalkeepers’ responses regarding their anxiety level in an empty-goal situation were analyzed in a similar manner.
A two-way ANOVA was conducted to evaluate the effect of gender and practice on the average of the nine anxiety items that are presented at
Table 4. We did not find a significant difference in anxiety between women and men (f = 0.81,
p = 0.368, df = 1), practice did not appear to be correlated with the keepers’ anxiety (f = 0.85,
p = 0.430, df = 2), and there was no significant interaction between gender and practice (f = 1.97,
p = 0.144, df = 2).
A two-way MANOVA revealed the difference between female and male keepers across nine anxiety items (see
Appendix B) to be non-significant (f = 1.07,
p = 0.389, df = 9, 81). The effect of practice was found to be significant (f = 2.03,
p = 0.010, df = 18, 164). For instance, we can see in
Appendix B that practice was found to have a significant explanatory power with regard to anxiety concerning the ability of the substitute to arrive at the substitution area on time (f = 3.23,
p = 0.044, df = 2); respondents with no practice exhibited more anxiety than keepers who participated in weekly practices of empty-goal scenario. The opposite trend emerged in responses regarding the possible damage to personal statistics during empty-goal scenarios; keepers with less practice were found to be less anxious (f = 4.76,
p = 0.011, df = 2). As for the interaction effect, there was no significant interaction between gender and the amount of practice across the nine anxiety items (f = 0.86,
p = 0.624, df = 18, 164).
An additional two-way MANOVA was conducted to measure the effect of gender, participation in a national team, and the interaction of these factors on the anxiety exhibited by goalkeepers regarding empty-goal scenarios. Participation in a national team was shown to be a non-significant factor (f = 1.53, p = 0.150, df = 9, 83), gender was shown to be non-significant (f = 1.13, p = 0.346, df = 9, 83), and the interaction between gender and national team participation was shown to be near-significant (f = 1.79, p = 0.081, df = 9, 83).
We proceeded with an examination of goalkeeper responses regarding their sense of belonging within their team during empty-goal scenarios, as well as the keepers’ overall opinions regarding the empty-goal rule.
We used a two-way ANOVA to examine the effect of gender, practice, and the interaction of these factors on the goalkeepers’ sense of belonging and overall approval of the empty-goal rule. Gender was not found to be a significant factor in regard to the keepers’ sense of belonging during empty-goal scenarios (f = 0.23, p = 0.627, df = 1), practice was not found to be significant (f = 0.95, p = 0.387, df = 2), and the interaction between gender and practice also failed to reach significance (f = 0.16, p = 0.850, df = 2). A similar analysis was performed with respect to participation in a national team in the place of the practice variable; none of the variables was found to be significant or near-significant.
As for the general opinion of goalkeepers regarding the empty-goal rule, gender was found to be near-significant (f = 2.86,
p = 0.094, df = 1), and we can see in
Table 5 that the female goalkeepers’ level of approval was 6.10 as compared to the male keepers’ overall opinion of 5.14. Practice was found to be positively correlated with the keepers’ opinion regarding the empty-goal rule (f = 3.56,
p = 0.032, df = 2), and the interaction between gender and practice was not found to be significant (f = 0.02,
p = 0.998, df = 2). A similar analysis was performed with participation in a national team in the place of the practice variable; none of the variables was found to be significant or near-significant.
We ran several linear regressions with average confidence, average anxiety, and overall approval of the empty-goal rule as the response variables.
Regressions results shown in
Table 6 provided a robustness check of the previously mentioned findings. We can see in
Table 6 that male goalkeepers are near-significantly more confident regarding their abilities in an empty-goal situation as compared to female keepers and that the amount of weekly practice is not associated with this confidence (model 1). Participation in a national team did not have any explanatory power with respect to confidence (model 2). These results correspond with the confidence averages presented in
Table 3.
A new finding is that more experienced goalkeepers were found to be near-significantly less confident in an empty-goal situation than less experienced goalkeeper (model 1). Additionally, a sense of belonging was found to be a significant predictor of confidence in an empty-goal situation (model 2).
Similar to results in
Table 4 regarding the keepers’ level of anxiety, the two regression models did not reveal any significant predictors of anxiety. As for overall opinions regarding the empty-goal rule, the obtained results suggest that less experienced goalkeepers are more positive about the empty-goal rule (near-significant result, model 1), that the amount of weekly practice of empty-goal scenarios is positively correlated with keepers’ approval of this new rule (significant in both models), and that female keepers held a more positive opinion regarding the empty-goal rule as compared to male keepers (model 2). Additionally, we can see that anxiety regarding the empty-goal rule is negatively correlated with approval of the rule (near-significant result, model 2).
Interestingly, we can see in
Table 6 that the sense of involvement within the team during empty-goal play is a strong predictor of the keepers’ overall opinion regarding this rule (model 2). We ran several additional models to test whether participation in the World Handball Championship (27 elite-level goalkeepers in our sample) has a unique exploratory power for confidence, anxiety, sense of belonging, or overall approval of the empty-goal rule; none of the models provided significant or near-significant results.