Friday, November 6, 2015

Blog #10-Credibility & Ethics

Essential Question: What primary concerns exist in ethics, validity and reliability in AR? How are you managing these concerns (or how will you) within your study?


Internal validity deals with the question of how research findings match reality. (Merriam & Tisdell, 2015, pp. 242) To measure validity, Lincoln and Gruba (1985) list credibility, transferability, dependability and confirmability as the main categories to assess the validity of a study. Credibility means that the study seems to be true. Transferability is that a study can be applied to another situation. Dependability refers to the ability of a study to be repeated with similar results. Confirmability means that the results are due to the respondents and they were not greatly influenced by the researcher.


In qualitative research “data do not speak for themselves; there is always an interpreter, or a translator” (Merriam & Tisdell, 2015, pp. 242). This means that no matter what we do as a researcher our opinion will always be present. “One cannot observe or measure a phenomenon/ event without changing it” (Merriam & Tisdell, 2015, pp. 242). Even when faced with similar situations people usually have different accounts.


According to Merriam and Tisdell (2015), some ways to increase the validity of a study include: triangulation, respondent validation, adequate time with data, peer review, audit trail, thick descriptions, and maximum variation. If you can get feedback from other sources either through reading the research or talking with others you can filter out some of your opinions. However, “the important feature of triangulation is not the simple combination of different kinds of data, but the attempt to relate them so as to counteract the threats to validity identified in each" (Brown, 2005).  This helps you to gain other perspectives that you may not have seen otherwise. Spending time collecting and analyzing data coupled with thick descriptions can help others to see where you are getting your conclusions. “If the findings of a study are consistent with the data presented, the study can be considered dependable” (Merriam & Tisdell, 2015, pp. 252). This also means that the researcher needs to explain their position so others will know where possibles biases may come into play.


When researching there will always be multiple opinions on what is allowed and what is not. The ethics behind the research is not black and white. There will always be gray areas. I think of biggest concern for my study will be confidentiality and anonymity. With using Twitter my students will be out in the online world. This means that I am not the only one who sees their tweets. I decided to look into the ethics of using Twitter to see if there was anything out there. There wasn’t much but I did come across the suggestion “that scientists never reveal screen names and make research objectives publicly available” (Moyer, 2014). My students are aware that others can see what they are doing and they know that they are posting to Twitter to see if it helps them out in class. They are aware of the basics but not the specifics. I think I will just have to be cautious in writing my conclusions to make sure that I think about their anonymity as I go.


This week I stumbled across another cartoon that made me chuckle. It went right along with my research and the ethical concerns that could come up.
twittercartoonnewyorker.jpg
https://epriego.wordpress.com/2014/05/27/twitter-as-public-evidence-and-the-ethics-of-twitter-research/


References


Brown, J. (2005). Questions and answers about language testing statistics: Characteristics of sound qualitative research. JALT Testing & Evaluation SIG Newsletter, 9(2), 31-33. Retrieved from http://jalt.org/test/bro_22.htm


Lincoln, YS. & Guba, EG. (1985). Naturalistic Inquiry. Newbury Park, CA: Sage Publications.


Merriam, Sharan B.; Tisdell, Elizabeth J. (2015-07-06). Qualitative Research: A Guide to Design and Implementation (JOSSEY-BASS HIGHER & ADULT EDUCATION SERIES) Wiley. Kindle Edition.

Moyer, M. (2014). Twitter to Release All Tweets to Scientists. Scientific American, 310(6). Retrieved from http://www.scientificamerican.com/article/twitter-to-release-all-tweets-to-scientists-a-trove-of-billions-of-tweets-will-be-a-research-boon-and-an-ethical-dilemma/

4 comments:


  1. Sara- I like what you quoted here from (Brown, 2005), However, “the important feature of triangulation is not the simple combination of different kinds of data, but the attempt to relate them so as to counteract the threats to validity identified in each".  This is good to know. That cartoon is funny, made me laugh as well. :)

    ReplyDelete
  2. The data just doesn't speak for itself... it would be so much easier if it did. I think the hardest part of putting all the results together and going through data is the fact that my opinion will always be present and I'll have to look at the bias and try to eliminate as much as possible, maybe it won't be that bad though.

    ReplyDelete
  3. Sara,

    I like that you said that research ethics will always have gray areas. This is so true. Ethics depend on who is conducting the research. I also agree that confidentiality and anonymity are the biggest issues in ethics that we face as educators. Protecting our students should always be our number one concern. I think your project is good for showing students about how social media is visible by anyone, and that you are helping students learn a valuable lesson about their digital footprint and digital citizenship.

    ReplyDelete
  4. I think you'll do great with protecting your students in your conclusion. You're right that it'll be hard to keep students confidentiality and anonymity using Twitter. Great idea with looking into the ethics part!

    ReplyDelete