Question

In: Statistics and Probability

In applying t-tests to your data and you find that your data are very non-normal. How...

In applying t-tests to your data and you find that your data are very non-normal. How would you expect that non-normality to affect your results if corrected by family-wise error rate (FWER)? How about by false discovery rate (FDR)?

Solutions

Expert Solution

If the sample size in 2 samples gets large, then the t-test is valid as the type 1 error rate(FWER) is controlled at the given significance level(usually 5%) even when the random variable, X doesn't follow the normal distribution. It's because of the fact that t-test is based on the mean values of two samples and . And thus, by the Central Limit Theorem(CLT), the distribution of these data, in repeated sampling, converges to a normal distribution, regardless of the distribution of X in the population. Also the estimator that t-test uses for the standard error of the sample means is consistent regardless of the distribution of X and so it is unaffected by non-normality.

The t-test only assumes that the means of the different groups are normally distributed but it doesn't assume that the population is normally distributed.

By the central limit theorem, we have that the means of samples from a population with finite variance approach follows a normal distribution irrespective of the distribution of the population. Sample means are normally distributed as long as the sample size is at least 20(some say 30). For a t-test to be valid on a smaller sample size, the population distribution would have to be normal (or approximately normal).

However, the t-test is invalid for small samples from non-normal distributions as the type error rate deviates from the given significance level (5%) and false discovery rate(FDR: proportion of falsely rejecting the null hypothesis when in fact it is true out of total rejections) may increase. But the t-test is valid for large samples from non-normal distributions due to CLT.


Related Solutions

How can a T-test be useful and very applicable in social science? As you discuss T-tests,...
How can a T-test be useful and very applicable in social science? As you discuss T-tests, note the several different kinds of t-tests (independent samples, paired-samples, etc.) and in what research context (experiments, medical trials, etc.) they may be more appropriate?
You are applying for a data job, and your assignment is to analyze the following data...
You are applying for a data job, and your assignment is to analyze the following data set for monthly average temperatures at St Catherines in the programming language R. (a) Prepare some plots that visualize the data. (b) Find the appropriate time series model and fit the data. Explain your choice of parameters. (c) Make predictions for the next 3 years, plot these predictions. (d) Summarize your findings. The Data is provided below. Please show the R codes as well....
1) An assumption of non parametric tests is that the distribution must be normal: (a) True...
1) An assumption of non parametric tests is that the distribution must be normal: (a) True (b) False 2) ​One characteristic of the chi-square tests is that they can be used when the data are measured on a nominal scale: (a) True (b) False 3) Which of the following accurately describes the observed frequencies for a chi-square test?​: (a) They are always whole numbers. (b) ​They can contain fractions or decimal values. (c) They can contain both positive and negative...
How many types of tests are considered non-parametric data and briefly explain each
How many types of tests are considered non-parametric data and briefly explain each
Find the subgroup of d4 and the normal and non normal subgroups of d3 and d4...
Find the subgroup of d4 and the normal and non normal subgroups of d3 and d4 using u and v, u being the flips and v being the rotations.
How to understand and compute non parametric tests
How to understand and compute non parametric tests
Explain how the chi-square tests differ from parametric tests (such as t tests or ANOVA) with...
Explain how the chi-square tests differ from parametric tests (such as t tests or ANOVA) with respect to the hypotheses, the data, and the assumptions underlying the test.
show your work. Find the sin t and cos t for t =150o. You may need...
show your work. Find the sin t and cos t for t =150o. You may need to draw the reference angle first, but you only have to enter the sine and cosine values.
Write a report about Non-parametric Tests and refere to your references
Write a report about Non-parametric Tests and refere to your references
How are paired t-tests different from two-sample t-tests? 1. When are they used: population parameters 2....
How are paired t-tests different from two-sample t-tests? 1. When are they used: population parameters 2. Data types they are used on 3. Which sample statistics are used as estimates for the population parameter 4. Test statistics, including conditions and Minitab command
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT