| Tag | Value |
|---|---|
| file | Inferential_Statistics_vufsw-paired_samples-1130-en_vufsw-paired_samples-1130-en |
| name | vufsw-paired samples-1130-en |
| section | inferential statistics/parametric techniques/t-test/paired samples |
| type | schoice |
| solution | FALSE, FALSE, TRUE, FALSE |
| Type | test choice |
| Program | NA |
| Language | English |
| Level | statistical reasoning |
In an experiment, 20 randomly selected subjects received training to improve their mathematics skills. Their performance was measured before the training took place (measurement 1) and afterwards (measurement 2) on a scale of 1 to 10. A higher score means better performance. The results are shown below. A researcher wants to investigate whether the training increases the score and performs a one-sided test with alpha = 0.05.
| N | mean measurement 1 (before) | mean measurement 2 (afterwards) | mean difference | standard error (se) mean difference |
| 20 | 5.75 | 6.35 | 0.60 | 0.24 |
The researcher states that these are two independent groups, so a dependent t-test is required. Explain why.
Whether the measurements are dependent or independent has nothing to do with whether there is or there is not an association between the X and the Y. The researcher took two measurements in the same group of subjects. The measurements are NOT independent (if a person scores at the beginning high on mathematics skills he or she will also score high at the second measurement, regardless of the intervention), therefore the two measurements are not independent and THUS is a dependent t-test is required.
If you want to know more about two dependent means watch this clip
Levels of Difficulty
Easy
M&T Hypothesis testing: means
Default value
M&T Model selection
Default value