VanillaCream
Badges: 13
Rep:
?
#1
Report Thread starter 1 month ago
#1
hey,
I need help with putting my data into one-way ANOVA test. I did an experiment about antibacterial properties of extracts of cinnamon and cranberry. I tested them on the sizes of zones of inhibition created by E. coli and S. epidermidis, in comparison to furazidin and amoxicillin. And I'm really struggling to put the data into the test. Can someone help and explain how I should do it? I found a calculator online, but I don't know which data I should imput.
I have 4 sets of data for E. coli (with furazidin, amoxicillin, cinnamon and cranberry) and 4 sets for S. epidermidis (again with furazidin, amoxicillin, cinnamon and cranberry).
Should I put (picture below) firstly as treatment 1 the data for E. coli with furazidin, then E. coli with amoxicillin, then E. coli with cranberry and E. coli with cinnamon and then the same for S. epidermidis? Or firstly e.g. E. coli with cranberry as treatment 1, then S. epidermidis with cranberry as treatment 2, then calculate and repeat the same for all other substances?
Please I need help
Name:  anova.PNG
Views: 10
Size:  25.3 KB
0
reply
Jpw1097
Badges: 17
Rep:
?
#2
Report 1 month ago
#2
(Original post by VanillaCream)
hey,
I need help with putting my data into one-way ANOVA test. I did an experiment about antibacterial properties of extracts of cinnamon and cranberry. I tested them on the sizes of zones of inhibition created by E. coli and S. epidermidis, in comparison to furazidin and amoxicillin. And I'm really struggling to put the data into the test. Can someone help and explain how I should do it? I found a calculator online, but I don't know which data I should imput.
I have 4 sets of data for E. coli (with furazidin, amoxicillin, cinnamon and cranberry) and 4 sets for S. epidermidis (again with furazidin, amoxicillin, cinnamon and cranberry).
Should I put (picture below) firstly as treatment 1 the data for E. coli with furazidin, then E. coli with amoxicillin, then E. coli with cranberry and E. coli with cinnamon and then the same for S. epidermidis? Or firstly e.g. E. coli with cranberry as treatment 1, then S. epidermidis with cranberry as treatment 2, then calculate and repeat the same for all other substances?
Please I need help
Name:  anova.PNG
Views: 10
Size:  25.3 KB
Presumably you have several replicates for each variable tested?

You will need to do a separate analysis for E. coli and S. epidermis.
An ANOVA year will not tell you which variable is significantly different from the others, if there is a significant difference. It depends on what your data shows. If your data shows that the size of the zone of inhibition of cranberry and cinnamon is similar to the antibiotics, then you might just want to do an ANOVA to show that the size of zone of inhibition of cinnamon and cranberry are similar to the other antibiotics.

You may just want to do t tests, comparing the zone of inhibition of cinnamon and cranberry to each antibiotic individually, especially if the zone of inhibition is quite different between the antibiotics.

If you decide to use an ANOVA, put in the values for the size of zones of inhibition for each antibiotic and the cranberry and cinnamon. Do 1 ANOVA for E. coli and another for S. epidermis.
0
reply
VanillaCream
Badges: 13
Rep:
?
#3
Report Thread starter 1 month ago
#3
(Original post by Jpw1097)
Presumably you have several replicates for each variable tested?

You will need to do a separate analysis for E. coli and S. epidermis.
An ANOVA year will not tell you which variable is significantly different from the others, if there is a significant difference. It depends on what your data shows. If your data shows that the size of the zone of inhibition of cranberry and cinnamon is similar to the antibiotics, then you might just want to do an ANOVA to show that the size of zone of inhibition of cinnamon and cranberry are similar to the other antibiotics.

You may just want to do t tests, comparing the zone of inhibition of cinnamon and cranberry to each antibiotic individually, especially if the zone of inhibition is quite different between the antibiotics.

If you decide to use an ANOVA, put in the values for the size of zones of inhibition for each antibiotic and the cranberry and cinnamon. Do 1 ANOVA for E. coli and another for S. epidermis.
Yes, I had 5 trials for each of the variables for each of the bacteria.

Thank you so much!

Would you say then that t-test would be aa better fit? I was initially considering a t-test, but my teacher suggested ANOVA, so I decided to go with it. And if so, which values should I compare together? As you said to compare e.g. cinnamon to furazidin, should I then put e.g. values for cinnamon in E. coli and furazidin in E. coli, then for cranberry in E. coli and furazidin in E. coli and then the same with amoxicillin and repeated for S. epidermidis?

Thank you once again for help!
0
reply
Jpw1097
Badges: 17
Rep:
?
#4
Report 1 month ago
#4
(Original post by VanillaCream)
Yes, I had 5 trials for each of the variables for each of the bacteria.

Thank you so much!

Would you say then that t-test would be aa better fit? I was initially considering a t-test, but my teacher suggested ANOVA, so I decided to go with it. And if so, which values should I compare together? As you said to compare e.g. cinnamon to furazidin, should I then put e.g. values for cinnamon in E. coli and furazidin in E. coli, then for cranberry in E. coli and furazidin in E. coli and then the same with amoxicillin and repeated for S. epidermidis?

Thank you once again for help!
I think a t test would be better, because if there is a significant difference in the size of zone of inhibition between amoxicillin and furazidin, then the ANOVA will tell you that there is a statistical difference when you include the cinnamon and cranberry, but it won’t tell you anything about whether cinnamon and cranberry are significantly different.

Therefore I think it would be better to do a t test comparing the zone of inhibition of cinnamon with furazidin, cinnamon with amoxicillin, and then compare cranberry with furazidin, cranberry with amoxicillin, and do this for both E. coli and S. epidermis.

Another consideration which I don’t think you’ve taken into account is whether you should be using a parametric test (i.e. t test, ANOVA) or a nonparametric test. Parametric tests are used when your data is normally distributed, if it is not, you should use nonparametric tests. The nonparametric equivalents of the t test and ANOVA are the Mann-Whitney U test (for two groups) and the Kuskal-Wallis test (more than two groups).
0
reply
VanillaCream
Badges: 13
Rep:
?
#5
Report Thread starter 1 month ago
#5
(Original post by Jpw1097)
I think a t test would be better, because if there is a significant difference in the size of zone of inhibition between amoxicillin and furazidin, then the ANOVA will tell you that there is a statistical difference when you include the cinnamon and cranberry, but it won’t tell you anything about whether cinnamon and cranberry are significantly different.

Therefore I think it would be better to do a t test comparing the zone of inhibition of cinnamon with furazidin, cinnamon with amoxicillin, and then compare cranberry with furazidin, cranberry with amoxicillin, and do this for both E. coli and S. epidermis.

Another consideration which I don’t think you’ve taken into account is whether you should be using a parametric test (i.e. t test, ANOVA) or a nonparametric test. Parametric tests are used when your data is normally distributed, if it is not, you should use nonparametric tests. The nonparametric equivalents of the t test and ANOVA are the Mann-Whitney U test (for two groups) and the Kuskal-Wallis test (more than two groups).
Oh okay, I didn't really think about it. Then probably Mann-Whitney U test would be the best and I could do it in the way you suggested for t test, as the data is not normally distributed, I think.

Thank you so much!
0
reply
Jpw1097
Badges: 17
Rep:
?
#6
Report 1 month ago
#6
(Original post by VanillaCream)
Oh okay, I didn't really think about it. Then probably Mann-Whitney U test would be the best and I could do it in the way you suggested for t test, as the data is not normally distributed, I think.

Thank you so much!
I think a nonparametric test would be more appropriate for small sample sizes, as you have.
1
reply
X

Quick Reply

Attached files
Write a reply...
Reply
new posts
Back
to top
Latest
My Feed

See more of what you like on
The Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

Personalise

Do you have the space and resources you need to succeed in home learning?

Yes I have everything I need (74)
64.35%
I don't have everything I need (41)
35.65%

Watched Threads

View All