Tukey-Kramer Post Hoc Test after One-Way ANOVA in Excel

This video demonstrates how to conduct a Tukey-Kramer test (post hoc test) after a one-way ANOVA using Microsoft Excel. The Tukey-Kramer test is a conservative post hoc test (controls Type I Error rate) and is used when the sample sizes for each level are unequal. In this example, three groups are compared using a one-way AVOVA using the Data Analysis tools in Excel. In these tools, the one-way ANOVA is referred to as “ANOVA: Single Factor.” The results of this ANOVA are interpreted. The formula to calculate the Tukey-Kramer test is constructed using the “mean square within,” the standard error, the sample sizes, and the means. The q statistic is calculated for every pairwise comparison and compared to the critical value for the studentized range distribution for q. To identify the correct critical value from the table, the number of groups and the degrees of freedom error (within-subjects df) are needed. A statistically significant result occurs when the q statistic exceeds the critical value.

[@davebsn]

Thank you for your well-paced, step by step explanation of how to do this process. I found it very easy to follow along and replicate the steps for my own work.

[@kingjonny394]

a lucky break finding this! best tutorial for this kind of thing i've ever seen

[@girlieserrano3799]

You actually saved our life somehow. Thank you!

[@sarahmcnulty1083]

Wonderfully explained! Thank you for your help. Best explanation I have found.

[@SuperPintu2010]

explained very fundamentally and conceptually!!!. Thank you!!

[@girlieserrano3799]

Like I really REALLY appreciate your tutorial! It was a sudden blessing for me. I might be over reacting but it is true. I was almost giving up since we weren't able to find statistician in nearby places that would help us but woah. Suddenly this video appeared in my recommendation. I really really want to say thank you sir!

[@DrGrande]

You are quite welcome -

[@brookerramos]

Thank you! Very easy to follow along :) Hopefully my IA will score well now!

[@chrisblack6269]

Thank you so much!! This saved me hours and hours for my data analysis.

[@amisha4707]

Absolutely amazing work!

[@DariusDevlin001]

This is excellent!

[@tessbrennan-kane9091]

Super helpful, thanks for sharing!

[@DrGrande]

You're welcome!

[@aragawfentabil470]

Thanks a lot for video!

[@abdullahalhejeilan1035]

thank you so much for this

[@allaouiamine523]

Thank you for your efforts!
But, are there a formula to calculate directly q crit and p-value, or at least, an excel function to do that?
Thank you.

[@ranganayakinandanavanam7362]

Very clear explanation and I have completed Turkey test for my data :)

[@DrGrande]

Thank you!

[@antoniofernandes5751]

thank you for this video

[@siewwoontan153]

hi dr. todd, if i were to use tukey test for between groups, does the SE formula remains the same and i just replace MSw to MSb?

[@iamnowsick]

thank you so much!

[@tumz1880]

What does it mean if I reject the null hypothesis for all q values? Does it mean that all groups contributed to a significance difference in the raw result?

[@maggiechen9607]

thank you so much

[@emmalonergan1836]

If I wanted to do a Tukey with an alpha=.1, would the steps be the same? Would the only difference be to find a new q critical?

[@tomhunt5643]

does this work with equal sample sizes too?

[@RomajaneAbrenio]

THANK YOU!

[@DrGrande]

You're welcome!

[@adelineln5916]

Can one use the Tukey procedure if one has four treatments? Or is it only limited to three treatments?

[@llamaliammm]

How do u get the q table?

[@Zannet24]

You just google it, Studentized Range q table :)

[@adrianjosephsenerpida2423]

How about for aliens like me?

[@ViscariaASMR]

You should really time stamp when you actually start talking about the tukey test so I wouldn't have to watch 8 commercials to get there.