1. Definition of error
- All different values
- When cannot measure a value perfectly so automatically introduces uncertainty
- A typical value
- A value that lies within the values
1 of 20
Other questions in this quiz
2. What is reliability?
- The measure of how similar the sample mean is as an estimate to the population mean
- The measure of how different the sample mean is as an estimate to the population mean
- The measure of how the parameters of sample and population are the same
3. Which is not a feature of different samples?
- There is not a significant difference between the mean and the small sample
- 100 samples to 10 samples
- So the smaller sample has committed a Type 1 error
- There is a high significant value between the mean with the bigger samples
4. Which is not a feature of the analysis of variance?
- he answer it gives you is about whether the means are the same or not.
- ANOVA compares all the means with each other
- What cant can accounted for by the difference- UNEXPLAINED VARIANCE/ WITHIN
- Null Hypothesis- it is that all the means are the same.
- It uses one overall variance. This is what allows there to be a single error rate, rather than many.
- What can be accounted for by the difference- EXPLAINED VARIANCE/ BETWEEN
- ANOVA makes similarities between the explained and unexplained variances.
- Alternative Hypothesis- at least one of the means differs significantly from at least one of the others.
5. Which is not a feature of the F statistic graph?
- The shape of the curve is defined by the degrees of freedom. The F distribution takes two d.f. values
- F Test compares two means to see whether one is significantly different that the other
- F-TEST compares two variances to see whether one is significantly bigger than the other
- F distribution is bounded at 0, with all F-values having to be positive. Both Z and t can be either positive or negative, and the normal and t distributions go to infinity in both directions, this is only true
- When both d.f. values are small it is very right-skewed. When both are large it looks a bit like the normal distribution. When one is small and the other large you get intermediate forms.
- The right-skew in the F-distribution remains, even for very large samples. Both the t and normal distributions are symmetrical.