Describe the research question you are pursuing for your final project ( Effects of Social Media on Academic Performance of College Students – research proposal with details is attached ). When drawing a statistical conclusion, what would a potential Type I error be? A potential Type II error? Why is it important for researchers to understand these errors?
When responding to your classmates, discuss the extent to which you agree with the Type I and Type II errors they identified, and also provide your perspective on why understanding these errors is important for researchers.
To complete this assignment, review the Discussion Rubric document.

AFTER COMPLETING THE INITIAL POST, PLEASE ALSO RESPOND TO THE FOLLOWING TWO STUDENTS REGARDING THE SAME TOPIC!
STUDENT ONE:
For my final project, my research question involves social media use and stress. More specifically, my question is, does more time on social media lead to more stress? I will be asking questions pertaining to the amount of time a person spends on social media, and then ask them to complete the Perceived Stress Scale. All of those questions will be about a timeframe of both a week and a month. My hypothesis is that the more a person spends on social media, the more stress they will have in their lives.
While drawing statistical conclusions, both null and alternative hypotheses need to be specifically made and pointed out. My hypothesis is the alternative, while the null is that the two variables (social media use and stress) will not be associated with each other in any way (Banerjee et al., 2009; Rosnow & Rosenthal, 2013). A potential Type I error would be a false positive. That is, I would reject the null hypothesis that there is no association between social media use and stress, but the data resulted in the null being true. A possible Type II error would be a false negative that I failed to reject the null, and it ended up being false anyway (Banerjee et al., 2009). So in another way, I was right, but I failed to reject the null hypothesis. Since we are limited to conducting our study on our classmates, our data will not be representative of the population because of the small sample size (Banerjee et al., 2009). Once we are able to conduct our very own research in our fields, then we can remedy that issue by increasing our sample sizes that way the conclusions will be tailored to a much larger population making generalizing more possible.
References
Banerjee, A., Chitnis, U. B., Jadhav, S. L., Bhawalkar, J. S., & Chaudhury, S. (2009). Hypothesis testing, type I and type II errors. Industrial psychiatry journal, 18(2), 127-131. doi:10.4103/0972-6748.62274
Rosnow, R. L. & Rosenthal, R. (2013). Beginning Behavioral Research: A Conceptual Primer. (7th ed.). Boston, MA: Pearson Education, Inc.
STUDENT TWO:
My research question is trying to determine if there is a correlation between the use of cell phones and the amount or quality of sleep. Rosnow and Rosenthal discuss Type I and Type II errors. According to the authors, “Type I error implies that the decision maker mistakenly rejected the null hypothesis when it is, in fact true and should not have been rejected. Type II error implies that the decision maker mistakenly failed to reject the null hypothesis when it is, in fact, false and should have been rejected” (2013, p 221). If I reject the null hypothesis when it was true, then it is a Type I. If I do not reject the null hypothesis and it is false, then this is the Type II.
It is possible to misinterpret the information by suggesting there is not a correlation between the two variables. One possible consideration for having a smaller sample size is altering the significant data to see if something is valid, it should have a highly significant outcome. For example, Rosnow and Rosenthal state the alpha, which in many of their examples from Chapter 12, discuss a significance of 5%. In an unofficial webpage, someone suggested using .01 rather than the .05 significance we have typically saw from PSY510. Their reasoning is that if the relationship is indeed valid, it will highlight at the more significant outcome and reduce chances for a Type I error. If the data I receive says that the variables are statistically significant, I would not reject the null hypothesis unless I lowered the alpha to accommodate for a smaller sample size. Additionally, if I do not notice a statistically significant relationship, I would accept the null hypothesis. Therefore, it is important to approach the research study cautiously.
References
Rosnow, L. & Rosenthal, R. (2013). Beginning Behavioral Research: A Conceptual Primer (7th ed). Boston, MA: Pearson Education, Inc.