Test table
,T-TESTS
1/1
Data normally distributed:
To test [state hypothesis] we performed a
dependent/independent/one sample(s)
[choose one] t-test. The assumption of
Homogeneity of Variance was met. On
average, [name highest variable] (M = x.xx, SD
= x.xx) was lower/higher than [lowest
variable] (M = xx.xx, SD = x.xx). This
difference was/was not significant (Mdif =
xx.xx, t(xx) =xx.xx, p </= .xxx). The 95% CI
(xx.xx, xx.xx) does/does not cross zero. The
difference represents a small/ medium/
large-sized effect d = .xx. We can conclude...
, T-TESTS
1/3
Data are not normally distributed:
To test [state hypothesis] we performed a dependent/independent/one
sample(s) [choose one] t-test. The data was not normally distributed (z-score
skewness/kurtosis = xx.xx and xx.xx). Therefore the p-value may not be
reliable and more weight should be placed on the bootstrapped 95%
confidence interval that will be provided. On average, [name highest variable]
(M = x.xx, SD = x.xx) was higher than [lowest variable] (M = xx.xx, SD = x.xx).
This difference was/was not significant (Mdif = xx.xx, t(xx) =xx.xx, p </= .xxx)
and generalizes/does not generalize to the population (95% CI xx.xx, xx.xx).
The difference represents a small/medium/large-sized effect d = .xx.
CI: see next page
,T-TESTS
1/1
Data normally distributed:
To test [state hypothesis] we performed a
dependent/independent/one sample(s)
[choose one] t-test. The assumption of
Homogeneity of Variance was met. On
average, [name highest variable] (M = x.xx, SD
= x.xx) was lower/higher than [lowest
variable] (M = xx.xx, SD = x.xx). This
difference was/was not significant (Mdif =
xx.xx, t(xx) =xx.xx, p </= .xxx). The 95% CI
(xx.xx, xx.xx) does/does not cross zero. The
difference represents a small/ medium/
large-sized effect d = .xx. We can conclude...
, T-TESTS
1/3
Data are not normally distributed:
To test [state hypothesis] we performed a dependent/independent/one
sample(s) [choose one] t-test. The data was not normally distributed (z-score
skewness/kurtosis = xx.xx and xx.xx). Therefore the p-value may not be
reliable and more weight should be placed on the bootstrapped 95%
confidence interval that will be provided. On average, [name highest variable]
(M = x.xx, SD = x.xx) was higher than [lowest variable] (M = xx.xx, SD = x.xx).
This difference was/was not significant (Mdif = xx.xx, t(xx) =xx.xx, p </= .xxx)
and generalizes/does not generalize to the population (95% CI xx.xx, xx.xx).
The difference represents a small/medium/large-sized effect d = .xx.
CI: see next page