I want to start a case control study. The case is a diabetic patient and the control is healthy individuals. The research is quantitative; I determine the bilirubin level between two groups. I want to select two control for each case the case to control is matching according to age sex and weight is it the correct way and how do I calculate sample size.
The best approach is to to do power analysis to estimate your sample size, you have to write your hypothesis, and based on that you decide what statistical test you will use. It should be one of the inferential statistics. so you need to determine the following: alpha {standard to be .05}, power [standard to be .80], effect size {small, moderate, or large, each test has its own value, you can find these values in the net}. Then download free programs to calculate the sample size such as G. power.
I find your question somewhat odd, but as a general rule of Thumb try to get at least 30 cases and if you want 2 controls for each then you need at least a sample of 90 individuals, I suggest your to read about sample sizes of 30 and confidence intervals, as well as t and z probabilities
If your outcome is quantitative (bilirubin level), you first want to establish the meaningful threshold or difference in bilirubin levels that you would want your study to have power to detect between the cases and controls (E.g. 10% diff or 20% diff). OR you can power your study to detect a small, moderate or large effect size. You can base your thresholds and also find the variability in the measure (Std Dev) from previous literature. GPower or Sample Power (from IBM SPSS Inc) are good softwares to use. I hope this was helpful.
Most articles/notes on sampling for case-control studies usually give you instructions for sample size based on the ODDs ratio. but sinc you are using a quantitative outcome, you can select the option in the software that allows you to calculate sample size based on mean differences between groups. (e.g paired t-test)
Raid & Samannaaz offer useful advice with regard to doing a power analysis. I would think of this in terms of using a t-Test as a way to compare your two groups, but you want a large enough sample that your test will have sufficient "power" to detect a meaningful result. In particular, too small a sample can easily lead to a false-negative, so that you fail to detect an effect that would be present with a larger sample..
The best approach is to to do power analysis to estimate your sample size, you have to write your hypothesis, and based on that you decide what statistical test you will use. It should be one of the inferential statistics. so you need to determine the following: alpha {standard to be .05}, power [standard to be .80], effect size {small, moderate, or large, each test has its own value, you can find these values in the net}. Then download free programs to calculate the sample size such as G. power.
To do power analysis to estimate your sample size, you have to write your hypothesis, and based on that you decide what statistical test you will use. It should be one of the inferential statistics. so you need to determine the following: alpha {standard to be .05}, power [standard to be .80], effect size {small, moderate, or large, each test has its own value, you can find these values in the net}. Then download free programs to calculate the sample size such as G. power.
- Badges
- Science topic
- Similar topics
- Medicine
- Disease
- Metabolic Diseases
- Glucose Metabolism Disorders
- Diabetes