Methodology Podcasts

Effect size is a population parameter. It describes the population. In a sample, the ideal one would be zero if there are no differences between the two groups. It would increase either in a positive or a negative direction as either p1 becomes much better than p2 in the positive direction, or p2 is much better than p1 in the negative direction.

If I randomly take a subject from the p1 population and I randomly take a subject from the p2 population, what’s the probability that the p1 subject is better than the p2 minus the probability that p2 is better than p1? If that number is 0, that difference between them, and that’s what this idea is, that number is 0, that’s going to be because there really isn’t any overall difference between the two populations. 

To learn more from Dr. Helena Kraemer listen to the podcast episode below.

Part 1

Part 2

Click here for more Research Methodology Podcasts

There is a definition of the P-value in every statistical textbook. Also there is a controversy in one way or another and it has been going on at least for the last 20 years. First I’m going to tell you what a P-value is not. A P-value is not the probability that your theory false. It is not the probability that the null hypothesis is true. It has nothing to do, really, with your hypothesis. It is a sign of how well you’ve done your research, major determinants of the P-value. The P-value being a statistic that you compute from the data in your study. The main influence on the P-value first of all, is the sample size. Second, the reliability of the measures that you use, the quality of your research design, the choice of analysis, you make. The fidelity with which you actually execute your research design, and how well you execute your analysis in the end. So, it all primarily has to do with the quality of the research.

To learn more from Dr. Helena Kraemer listen to the podcast episode below.

Part 1

Part 2

Click here for more Research Methodology Podcasts

What do you do when you have a set of observations, you don’t know what the distribution is, but you have a certain question, for example I want to test between two populations and I don’t know what the distributions in either are? What you do is, if you have let’s say 10 observations in the sample, drop one and then compute statistic, put it back, drop a second one computer statistic. At the end if your 10 observations, you have 10 what are called jackknife estimates. Each one is obtained by dropping one of the original observations. Though the jackknife method actually worked very well for a lot of cases and it was the method of choice for quite a long time, it also had some problems. It didn’t work very well with small samples and it didn’t work if you were dealing with data where there were a lot of ties. But then when we got to about the 80s, and the 90s, that’s when the bootstrap method came out. 

To learn more from Dr. Helena Kraemer listen to the podcast episode below.

Click here for more Research Methodology Podcasts

If you’re doing exploration, if you’re doing hypothesis testing, you have to have an priori hypothesis. a hypothesis has a rationale and justification, often drawn from previous hypothesis-generating studies.

You set up a plan, who are you going to sample? What are you going to measure, how you are going to measure it, how you’re going to analyze it, what the power calculations are, and then you do it as planned. That’s hypothesis testing. Hypothesis generation, or exploratory methods, basically means you don’t really have a specific plan in mind. You have sort of a general idea of the kind of thing you’re looking for, and you’re going to go out and you’re going to see what you see. And what you see is going to condition what analyses you’re going to do, and the results of those analyses are going to generate new questions that you’re going to do more analyses on, you can spend a lot of time doing this, and I will tell you this, and that is hypothesis testing is hard work.

To learn more from Dr. Helena Kraemer listen to the podcast episode below.

Part 1

Part 2

Click here for more Research Methodology Podcasts

Fidelity means in a research study that you do the study exactly as you plan to do the study. It’s really crucial when you’re planning a study to set up what your sampling criteria are, what population you want your results to apply to, what your inclusion, exclusion criteria are. Where are you going to get the subjects for your study? What measurements are you going to get, when you’re going to get them, how you’re going to get them, by whom you’re going to get them. All of this is part of design and, once the study starts, nothing should change. It should be done exactly as you plan to do it. Fidelity is that the researchers do the study as they plan to do it.

To learn more from Dr. Helena Kraemer listen to the podcast episode below.

Part 1

Part 2

Click here for more Research Methodology Podcasts