MODULE 3 DISCUSSION

Provides substantive responses to the below discussions:

1. Add value to the discussions below

2. By presenting alternative points of view,

3. Asking thought-provoking questions,

4. Engaging in respectful critique, and/or adding additional support to the discussion viewpoints

5. Post length (125-150 words each)

This is Tanya’s Discussion

This discussion post will discuss my opinion about type 1 and type 2 errors in research. I will explain which would most likely taint a research finding. Additionally, I will consider the significant test and effect size.

A Type 1 error in research, is considered to be when a hypothesis is rejected but is determined true. A Type 2 error in research, is when the hypothesis is accepted but determined to be false (HIV/AIDS & Drug Treatment). In my opinion, both are serious issues when conducting research. A Type 2 error in research could be misleading to the participant. In other words, the participants may believe that the researcher found a cure to problem. When actually, the researcher did not find a cure or resolution to the problem.  In the matter of a Type 2 error, the researcher could possibly make a major mistake, (A negative test result for HIV/AIDS).

If you had to report only one, which one would you report and why? If I had to only report one, I would report the Type 2 error.  Based on the example I used in this discussion. Providing someone with a false negative test result, could result in the participant (s) affecting other people. 

Explain whether or not you would report confidence intervals with either one. I would report confidence intervals in both research methods. Providing the information about confidence intervals, would provide information about uncertainty within the research findings.

This is Heather’s Discussion

In your opinion, what would most likely taint a research finding, a type 1 or type 2 error?  Be sure to explain your response.

            There are two major types of errors that occur in quantitative research, type 1 and type 2. Reviewing this week’s readings, Huck (2012) describes a type 1 error to mean the null hypothesis is rejected when it is correct, and type 2 error to mean that the null hypothesis is not rejected when it is false. This can be confusing, as it seems that we are talking in double negatives, and this is because the researcher is focusing on the null hypothesis, which is assuming there is no relationship between the variables.  

I am struggling to determine which error would be worse, as they are both errors and create frustrating results with the research. I believe both would taint the research findings, however, a type 2 error would mean that my hypothesis is wrong when it is actually correct. Should I receive these results with a study, I would want to re-test and review the results again in attempt to prove my hypothesis. 

Consider the significant test and effect size. If you had to report only one, which one would you report and why? Explain whether or not you would report confidence intervals with either one.

            As a researcher and a counselor, I understand ethics and believe that all research and data should be transparent. I have been instructed to document all findings and to list all steps in the research process so that the study may be duplicated by others in the future. However, if I needed to choose to report the significant test or the effect size, I believe I would report the effect size. 

Gross-Sampson (2019) describes the effect size to measure the difference of two groups. And as last week’s homework, reviewing the Pearson Correlation, we studied the significant test, or the p-value, I was inclined to remark that I would choose to report the significant test. However, Sullivan and Feinn (2012) state that effect size is independent of sample size, yet significance is dependent upon both sample size and effect size. For this reason, I would report the effect size if I was in need of choosing to report only one. 

I am interested in seeing other’s results of this discussion and hearing opinions about the topic.