Hello Everyone,
I could use some serious assistance. It was brought to my attention that use of individual T.tests [and F-tests] are not enough to discover statistical analysis as it increases the likelihood of my Type 1 error.
My experimental design is the following:
Ca Imaging
Factor 1: Gender Males vs. Females
Factor 2: Wild-Type (Control) vs. Gene-Knockout Animals (non-Control)
Factor 3: Treatment A (Control) vs. Treatment B (Non-control)
I measure three parameters (So I'd repeat this test 3 times) as well, and the comparison of percentage change from Parameter A to B ( Tx B *100 / Tx A)
Electrocardiography
Factor 1: Gender Males vs. Females
Factor 2: Wild-Type (Control) vs. Gene-Knockout Animals (non-Control)
Factor 3: Treatment A (Control) vs. Treatment B (Non-control 1) vs. Treatment C (Non-control 2) vs. Treatment D (Non-control 3)
I measure four parameters (So i'd repeat this test 4 times) , and I measure the percentage change from Treatment A to B; Treatment C to A; and Treatment D to A.
Hello there,
I am not sure I understand [1] Factor 3 components in both cases, and [2] The number of repetitions (replications).
I am going to give you what a typical design (factorial) should look like for a three way-analysis of variance (ANOVA), which means you have three factors:
Factor 1: Gender (male/female)
Factor 2: Genetics make up (wild/modified)
Factor 3: Treatment (e.g. feeding the animals three types of feed/drug, etc)
Now your design will look like the Table 1 in the attached file, where your study animals from both groups (male versus female; wild versus modified) should be subjected the treatment.
Your three-way ANOVA will contain the following components of variability:
- Gender (how your parameters vary between gender)
- Genetics (how your parameters vary between the wild and Gene-knock out)
- Treatment (how different the parameters are from the different levels of treatment)
- Gender*Genetics (how the different gender groups responded to the gene modification)
- Gender*Treatment (how the different gender responded to the different treatments)
- Genetics*treatment (how the gene modification is varying with the treatments)
- Gender*Genetics*Treatment (how the different gender group with different gene-make up are responding to the different levels of the treatment)
- Error (variability due to random error)
- Total (total variability in the data set)
Based on the significance of the ANOVA, you can proceed to the mean separation for each component using Tukey's HSD or other techniques. I hope this helps.
Tadesse Fikre Teferra
Hello Dr. Teferra,
Thank you so much for taking the time to assist me with my question.
I will attempt to perform this analysis and interpret the differences to the best of my ability.
If I have any further questions, may I reach out to you directly for further assistance?
Sincerely,
Ronen Ostro
Of course! I will be happy to help if I understand the data and research.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA