Skip to main content

Table 3 Main statistical analyses and interpretation of statistics

From: A systematic review of the validity, reliability, and feasibility of measurement tools used to assess the physical activity and sedentary behaviour of pre-school aged children

 

Relative or Absolute Validity/reliability?

Weak

Moderate

Good

Correlations (r)

 Pearson’s

Relative

< 0.60

0.60–0.79

≥0.80

 Spearman’s

Relative

< 0.60

0.60–0.79

≥0.80

 Kendall’s

Relative

< 0.60

0.60–0.79

≥0.80

 Intraclass correlation coefficient

Absolute

< 0.60

0.60–0.69

≥0.70

Linear Regression (% variance explained by the measurement tool)

Relative

< 60%

60–79%

≥80%

Root mean squared error

Absolute

*

*

*

Bland Altman (mean difference, limits of agreement, bias)

Absolute

*

*

*

Kappa (r)

Absolute

< 0.60

0.60–0.69

≥0.70

Area under the receiver operating curve (AUC-ROC)

Relative

< 0.70

0.70–0.79

≥0.80

Feasibility (% of usable data)

< 60%

60–79%

80–100%

  1. (References: ([23, 26, 28, 29, 35, 36, 46–53]))
  2. *Depends on the units of measurement