Table 2. Results of moderator analyses for criterion-related validity correlation coefficients across the 20-m shuttle run test protocols potentially affected by moderator effects†
| Moderator |
Effect |
K |
N |
rc |
rp |
95% CIa |
95% CVb |
% variancec |
Q statistic |
| Between-study analyses |
| Sex of participants |
| Légerd |
Men |
24 |
782 |
.80 |
.88 |
.81-.94 |
0.59-1.00 |
18.35 |
124.53* |
|
Women |
13 |
475 |
.74 |
.81 |
.70-.92 |
0.50-1.00 |
21.56 |
55.14* |
| Eurofite |
Men |
6 |
125 |
.61 |
.68 |
.49-.87 |
0.41-0.95 |
57.29 |
5.44 |
|
Women |
4 |
98 |
.67 |
.75 |
.53-.97 |
0.38-1.00 |
32.74 |
10.00* |
| Age of participants |
| Légerd |
Children |
28 |
1,335 |
.72 |
.78 |
.72-0.85 |
0.50-1.00 |
22.30 |
113.74* |
|
Adults |
20 |
887 |
.86 |
.94 |
.87-1.00 |
0.72-1.00 |
12.72 |
160.02* |
| Eurofite |
Children |
7 |
143 |
.61 |
.68 |
.52-.84 |
0.50-0.86 |
76.18 |
2.66 |
|
Adults |
4 |
135 |
.71 |
.79 |
.56-1.00 |
0.43-1.00 |
24.49 |
15.00* |
| Level of VO2max |
| Légerd |
Low |
22 |
1,181 |
.75 |
.82 |
.74-.89 |
0.49-1.00 |
13.27 |
167.60* |
|
High |
23 |
895 |
.80 |
.88 |
.80-.95 |
0.62-1.00 |
18.84 |
115.48* |
| Eurofite |
Low |
5 |
108 |
.69 |
.77 |
.59-.94 |
0.41-1.00 |
36.29 |
10.68* |
|
High |
6 |
170 |
.64 |
.71 |
.52-.91 |
0.45-0.98 |
48.32 |
7.81 |
| Within-study analysis |
| Number of predictorsf |
| Légerd |
One |
8 |
742 |
.70 |
.77 |
.68-.86 |
0.51-1.00 |
15.56 |
50.61* |
|
Few |
8 |
742 |
.80 |
.88 |
.80-.95 |
0.67-1.00 |
12.60 |
64.68* |
* p < 0.05
f Performance only score (“one”) or performance score plus other (“few”). † Because some studies mixed categories or some values were missing, the overall n for some categories is lower for some 20-m shuttle run tests.
eEurofit protocol starts at 8.0 km/h and increases 0.5 km/h each minute, but the second stage increases by 1.0 km/h (Council of Europe Committee for the Development of Sport, 1988)
dLéger’s protocol starts at 8.5 km/h and increases 0.5 km/h each minute (Léger et al., 1984, 1988)
a95% confidence interval
b95% credibility interval
cPercentage of variance accounted for by statistical artefacts including sampling error and measurement error of the 20-m shuttle run testNote. K, number of rs; N, total sample size; r
c, overall weighted mean of r corrected for sampling error only; r
p, overall weighted mean of r corrected for sampling error and measurement error of the 20-m shuttle run test; VO
2max, maximal oxygen uptake
