Fisher's least significant difference

Topic 6 DQ 2

How does the Fisher’s least significant difference test differ from Tukey’s honestly significant difference test? Provide examples of when you would use each

Luis

Re: Topic 6 DQ 2 Professor and Classmates,

Both are post hoc tests used to determine which differences between pairs of group means are significant. The Fisher's least significant difference test, also called the protected t test, is the most liberal post hoc test allowed, it has the lowest critical values acceptable, which increases the power to detect an effect or mean difference between two groups and it does not require that the number of participants in each group be equal.

The Tukey's Honestly Significant Difference test is a more conservative test where the critical levels are larger for each pairwise comparison. Both tests produce the same results and would be used when making a decision to either retain or reject the null hypothesis.

An example of when to use the Fisher's LSD test could be to gather data on the level of stress experienced by high school students in each of the four classes; freshman, sophomore, junior, and senior class, and how it affects learning, since each class would not have an equal number of students attending/enrolled.

An example of when to use the Tukey's HSD test could be to gather data on a random sample of participants exercising in a gym to determine if their exercise routine would be affected if distracted by having music or videos playing or absent during their exercise period. More specifically to gather data on length of time spent exercising, weight loss, or mood/focus during the exercise period observed.

Terry

Re: Topic 6 DQ 2 Turkey’s honestly significant difference test is the most conservative while Fisher’s least significant difference test is the most liberal. Fisher’s test is also called the protected t test, the number of people in each group could be different. The Fisher’s test has the lowest critical values that are acceptable, this increases the power to see an error or mean difference between groups. Fisher’s least significant difference test can be used to compare means of different groups. For example; if you want to see what group learned things faster you could compare by testing groups after 2 days, then 4 days and finally after 6 days to see which has the highest mean. The Turkey’s test is used for more pairwise comparisons. For example; this test can be used to compare a large number of means, using more groups with larger numbers of people in groups. For this test you could use a complete school and compare the differences in age and study habits.

Patricia

Re: Topic 6 DQ 2

Unlike the conservative Tukey’s honestly significant difference test, the Fisher’s LSD test, is the most liberal post hoc test allowed in published research studies (Privitera, 2015). It also has the lowest critical values acceptable, which increases the power to detect an effect or mean difference between two groups and the test does not require that the number of participants in each group is equal (Privitera, 2015). The Fisher LSD is used to compare the individual error rate and number of comparisons to calculate the simultaneous confidence level for all confidence intervals. On the other hand, the Tukey test was designed to allow one to make all of the pairwise comparisons. The Tukey test is often used if data analysis indicates that there may be some interesting information that cannot be analyzed through the preplanned test or when you need to compare large numbers of means.

Privitera, G. (2015) Statistics for

Mary

Re: Topic 6 DQ 2

Post hoc tests are used in statistics to evaluate the differences for all likely pairs of group means. Two of these posts hoc tests are Fisher’s least significant difference (LSD) test and Tukey’s honestly significant difference (HSD) test. Tukey’s HSD is the most conservative, and Fisher’s LSD is the most liberal that is allowed in published research studies (Privitera, 2015). Fisher’s LSD is said to have the most power in detecting an effect, which is because of its low critical values. It is also not necessary for the number of participants to be equal in each group (Privitera, 2015). An example of this test would be comparing the differences between two means in order to show the LSD. On the other hand, with Tukey’s HSD, the critical values are larger because it is more conservative. This test uses the student range statistic in its second step, which is used to find the critical value. An example of when to use Tukey’s LSD can be seen when a researcher wants to whether or not students learn more effectively with a constant background noise, as opposed to an unpredictable sound or no sound at all.

References

Privitera, Gregory J. (2015). Statistics for the behavioral Sciences. Retrieved from https://viewer.gcu.edu