A colleague recently asked me a statistics related question, one that would be shortened to "comparing means between two groups". I directly suggested a t-test. She agreed, but then asked another colleague who suggested an ANOVA instead. I don't understand why one would use ANOVA instead of a t-test. Any ideas?
To compare the means with two independent samples, it is preferable to use the Welch test. It is also known as the t-test for not equal variances case and it keeps you away from many problems related with heteroscedascity scenarios.
Please read the following article:
https://www.researchgate.net/publication/262687158_Se_necesita_la_prueba_de_Student_para_dos_muestras_independientes_asumiendo_varianzas_iguales
best regards
Jorge Ortiz
Thank you! Would there be any exceptions to this rule? Exceptions in which you would use ANOVA to compare two groups?
comparison between two means T-test will be used and ANOVA to caparison between more than 3 groups...
There is a thin line of demarcation amidst t-test and ANOVA, i.e. when the population means of only two groups is to be compared, the t-test is used, but when means of more than two groups are to be compared, ANOVA is preferred
When having unequal variances in your two groups, ANOVA is not the method of choice. The t-test will still work (but now with unequal variances).
ANOVA is simply a generalization of the t-test for more than 2 groups. If you do ANOVA on 2 groups you have done a t-test though the statistic may be presented as an F-statistic (which reduces in this case to a t-test). In other words, there is nothing to be gained by doing an ANOVA. But it is not necessary or better than a t-test.
To compare the means with two independent samples, it is preferable to use the Welch test. It is also known as the t-test for not equal variances case and it keeps you away from many problems related with heteroscedascity scenarios.
Please read the following article:
https://www.researchgate.net/publication/262687158_Se_necesita_la_prueba_de_Student_para_dos_muestras_independientes_asumiendo_varianzas_iguales
best regards
Jorge Ortiz
Research Welch's t-test for comparing two independent groups: An Exce...
I agree with Jorge Ortiz Pinilla
Welch's t-test is preferred even if you have equal sample sizes and variances.
If you want to compare the differences in outcomes between Student's t-test and Welch's t-test, please check out my calculator.
With Welch statistics you can directly test one side hypotheses and obtain one side confidence intervals if you need them. This is very useful in many instances and anova does not help so easily.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA