.

Miner

TS Contributor
#3
I recommend using Dunnett's post-hoc test. It will test drugs A and B against the placebo reducing the total number of tests.
1660566882649.png
1660566833073.png
 
#4
Oneway analysis of variance with a post-hoc test seems logical.

With kind regards

Karabiner

Thank you so much for your reply!

Just want to make sure i understand, the reasoning behind choosing these methods is:

1) I will choose one way ANOVA since i only have 1 independent variable (even though it has 3 levels)

2) i will go for ANOVA first because i need to identify whether or not the means of my 3 groups differ
So i will have a null hypothesis: All group means are the same, then when i reject my null hypothesis i go for post hoc

3) i will go for a post-hoc test to identify which mean is different?

-Is there a specific post-hoc test i should go for?

-Also, the question does not ask for the difference in mean average, i just assumed that ANOVA would be the best?
 

Miner

TS Contributor
#5
Thank you for your suggestion

I am just wondering how i would input that data in SPSS? How would i be measuring Drug A versus Placebo and then Drug B versus Placebo?

Since my initial chosen test would be one way ANOVA to measure my three groups (Drug A Drug B and Placebo)
Sorry, I do not use SPSS.
 

Miner

TS Contributor
#6
Also, the goal of my test is to find out which drug had the least negative effect on reaction time (the data in the above table is reaction time measured in seconds)

So how would the Dunnett's post hoc test help me identify the Drug?
The test tells you that Drug A significantly increases reaction time vs the placebo, and that Drug B does not significantly increase the reaction time vs. the placebo. Therefore, Drug B appears to be the answer.
 

Miner

TS Contributor
#7
Most of these charts just provide summary statistics of the three drugs. The key elements are:
1) 1-way ANOVA table. This shows that there is at least one contrast between the three drugs that is statistically significant.
1660596318882.png

2) Multiple comparisons using Bonferroni post-hoc test. This shows that Drug A is statistically different from both Drug B and Drug C.
Drug B is statistically different from Drug A, but NOT from Drug C/Placebo. Drug C/Placebo is statistically different from Drug A, but NOT from Drug B.
1660596463872.png
There is a lot more information in the screen shots, but this is the more relevant information to your questions.
 
#8
Most of these charts just provide summary statistics of the three drugs. The key elements are:
1) 1-way ANOVA table. This shows that there is at least one contrast between the three drugs that is statistically significant.
View attachment 4292

2) Multiple comparisons using Bonferroni post-hoc test. This shows that Drug A is statistically different from both Drug B and Drug C.
Drug B is statistically different from Drug A, but NOT from Drug C/Placebo. Drug C/Placebo is statistically different from Drug A, but NOT from Drug B.
View attachment 4293
There is a lot more information in the screen shots, but this is the more relevant information to your questions.


Thank you for taking your time to go through all of this. I am still finding this a bit difficult to understand. Does this mean that Drug B still has the most detrimental effect on the reaction time?
And is the bonferroni the right post hoc test to use to help me identify the drug? Or should i use something else?


For the graph 1, how were you able to identify that not all drugs had the same mean? Was it the degrees of freedom or F value?
 

Miner

TS Contributor
#9
The significance (Sig.) level (or p-value) being less than 0.05 (typical alpha risk) indicates statistical significance. You stated in an earlier post that higher response times were detrimental. Drug A is the only drug that is statistically different from the placebo AND it has a higher mean. Therefore, it is the only drug that is more detrimental than the placebo.
 

Miner

TS Contributor
#10
1) Means and Standard deviation: can you help explain what these numbers mean?

View attachment 4294
Means = Averages
Standard Deviation is a measure of the variation in the sample
N = sample size

2) Does this table mean that my means for each drug are not the same?

View attachment 4295
This is an ANOVA (ANalysis Of VAriance) table. It is a hypothesis test for testing if there is a statistically significant difference between the average of any one drug. The small significance level means that at least one drug is different from the other drugs.
This is the Bonferroni post-hoc test that I explained above.
4) Does this mean i have to use post hoc?
View attachment 4297
This is just a bar chart of the averages for each drug. You have already done the post-hoc as previously explained.