A total of 50 individuals of the species were collected monthly from sampling area for 13 months to study population dynamics of this bivalve. I have always understood that we need at least 30 to be able to reasonably expect a statistical analysis, but I heard lately that it was not enough! what do you think about this please? Is there a specific condition for determining the sample size?
50 randomly samples specimens should be enough, but as so often "the more the merrier".. if you had 100 specimens that would be much better, too, improving the overall data distribution. Just have a look at previous literature and sample sizes and how peers struggled (or not) in getting the statistical analyses right.
I suggest you should go for 100 sample, which would give a better result
Hello Hajar; Your question is about the required "statistical power" of your analysis. It would be helpful for you to do a bit of reading in an introductory statistical analysis source that clarifies some of the issues of the
1. range of variation in your samples
2. possible type I and type II errors in the kinds of dimensions you are considering.
An excellent source is.
Ruxton GD and N Colegrave. 2003. Experimental Design for the Life Sciences. Oxford Univ Press.
It is very important to get a clear idea so that your analysis actually measures what you hoped to measure. Best regards, Jim Des Lauriers
Hi Hajar, population dynamics involves several response variables (sex ratios, growth rate, etc). Since you did not mention what variable(s) is(are) of interest, a good practice would be checking the dispersion of your data using scatter or other type of plots. You can plot the mean of your variable with some measure of the dispersion (standard error, standard deviation) and see how much variation occurrs among your sampling months. A large dispersion in your data set may indicate low sampling size. Also, as mentioned above, you could also test for statistical power. Another example would be calculating biomass per sampling area (or effort) and see at which sample number the data reaches asymptote. There are other options too such as the calculation of intervals of confidence, but you need to identify your varaibles. I hope this helps.
Sample Size Rule
Sekaran (2013) wrote:
"Roscoe (1975) proposes the following rules of thumb for determining sample size:
1. Sample sizes larger than 30 and less than 500 are appropriate for most research.
2. Where samples are to be broken into sub-samples;(male/females, juniors/seniors, etc.), a minimum sample size of 30 for each category is necessary.
3. In multivariate research (including multiple regression analyses),the sample size should be several times (preferably 10 times or more) as large as the number of variables in the study.
4. For simple experimental research with tight experimental controls (matched pairs, etc.), successful research is possible with samples as small as 10 to 20 in size."
Reference
Sekaran, U., 2003. Research methods for business: A skill building approach. John Wiley & Sons.
Hello Hajar, it is important to obtain a minimum number of individuals (50-100) in a population dynamics study and it is also important to have a wide range of sizes. The case that you have few individuals due to the effect of fishing (or fishing gear) and I suppose that this causes that you can only obtain about 30 individuals per month. However, the results of the population that you are going to obtain are the ones that really exist. The growth results that you can obtain can be contrasted using the growth performance index OGP or Phy-Prime with other similar areas even if your data does not have a wide range of sizes. Apart from the FISAT for calculations you can use the TROPFISH -R which has 2 interesting routines on growth.
Greetings,
Marco A.
https://www.researchgate.net/project/TropFishR
- Badges
- Science topic