Skip to main content

Table 3 Comparison of scores between individuals with and without previous nutrition education; test-retest reliability; internal reliability analysis; results of Rasch analysis

From: The nutrition for sport knowledge questionnaire (NSKQ): development and validation using classical test theory and Rasch analysis

Section

Score (%): Nutrition education

Score (%): No nutrition education

Score p-level

Pearson’s Correlation R (p- level)

KR-20+

Dimensionality (Per < 5%)

Bonferroni adjusted p-level

Item fit statistics

*Overall (n = 89)

64. 65 (16)

52 (19)

<0.001

0.92 (<0.001)

0.0.87

19.3%

0.0005

Persons SD = 1.0271, Mean = −0.0938

Items SD = 1.324, Mean = 0.0724

Chi-Square probability = 0.000

**WeightManagement (n = 13)

77 (23)

62 (31)

<0.001

0.0.81 (<0.001)

0.62

2.8%

0.004

Persons SD = 0.62, Mean = −0.12;

Items SD = 1.06; Mean = 0.02;

Chi Square probability = 0.01 > 0.004

*Macronutrients (n = 30)

78 (15)

61 (16)

<0.001

0.0.81 (<0.001)

0.78

5.0%

0.002

Persons SD = 0.72, Mean = −0.11;

Items SD = 1.35; Mean = 0.05;

Chi Square probability = 0.0000 < 0.002

**Micronutrients (n = 13)

69 (23)

50.0 (30.8)

<0.001

0.76 (<0.001)

0.71

0.0%

0.005

Persons SD = 0.80, Mean = −0.21;

50 (31)

Items SD = 1.16; Mean = −0.36;

Chi Square probability = 0.0002 < 0.0004

**Sports Nutrition (n = 13)

54 (23)

46 (23)

<0.001

0.0.73 (<0.001)

0.63

1.7%

0.004

Persons SD = 0.57, Mean = −0.19;

Items SD = 1.39; Mean = 0.06;

Chi Square probability = 0.009 > 0.004

**Supplements (n = 12)

33 (33)

25 (25)

<0.001

(0.60.35 (0.6)

0.0.69

0.0%

0.004

Persons SD = 0.69, Mean = −0.13;

Items SD = 0.82; Mean = −0.20;

Chi Square probability = 0.02 > 0.004

**Alcohol (n = 8)

63 (31)

63 (25)

0.0.80

0.66 (<0.001)

0.0.51

0.0%

0.007

Persons SD = 0.65, Mean = −0.14;

Items SD = 1.1; Mean = −0.09;

Chi Square probability = 0.003 < 0.007

  1. *Normally distributed; scores reported as Mean (SD) **Non-normally distributed; scores reported as Median (IQR). CTT: Significant differences in groups scores indicate construct validity; Pearson’s r of 0.7 indicates adequate test-retest reliability; a KR-20 of 0.7 indicates adequate internal reliability. Rasch analysis: A non-significant Bonferroni adjusted chi-square probability indicates items are equally difficult across all participants (i.e. are good discriminators); a SD of 1 and Mean of 0 (for persons/items) indicate perfect fit to the Rasch model; a SD of >1.5 indicates the assumptions that difficult items are less likely to be answered correctly or well-scoring participants are more likely to answer questions correctly has not been met