Interval or Ratio as Scale of Measurement - Assignment Example

The paper "Interval or Ratio as Scale of Measurement" is a delightful example of an assignment on statistics. Repeated measure T-test would be the appropriate statistical analysis here because the current research consists of more than one set of samples (two sets in this case) but from the same group of participants.

• Assumptions in parametric tests:
  The assumptions for T-tests are interval or ratio as the scale of measurement, random sampling of the population, and normality of the distribution of scores (for each participant) in the population. The first two assumptions are met and here, we will particularly test Normality since this assumption is even more important when the sample size is small (less than 30) as in this case.

• Testing the assumption:

For CAFPP:

• We know that for a normal distribution the value of skewness and kurtosis are 0. However, we need to check if the score deviates significantly from 0. As a rule of thumb, |s| >1 & |k| >3 indicates serious non-normality. Here, both the values are well within range and therefore, the sample can be assumed to be normal (output 1).

Case Processing Summary

Descriptives

Statistic

Std. Error

CAFPP

Mean

480.6095

17.28538

95% Confidence Interval for Mean

5% Trimmed Mean

Median

Variance

Std. Deviation

Minimum

Maximum

Range

Interquartile Range

Skewness

Lower Bound

Upper Bound

485.6349

460.0000

6274.474

79.21158

260.00

603.00

343.00

120.0000

-.633

444.5528

516.6662

.501

Kurtosis

1.702

.972

Output 1

• This paper tells that the significance value of K-S is more than alpha 0.05 (Output 2) and indicates that the sample can be assumed to be normal.

Tests of Normality

Kolmogorov-Smirnov(a)

Statistic

Shapiro-Wilk

df

CAFPP

.186

21

a Lilliefors Significance Correction

Output 2

The graphical data (Output 3) suggest that although not a clear normal distribution is shown by histogram and stem plot can, however, be safely assumed to be approximately normal. Also, the points in the Normal Q-Q curve fall almost in a line.