Tuesday, September 5
Good work in class today with looking deep into research results and thinking through what went wrong in the study. Not too much was discussed today, but what we did talk about is very important. Today, we went over different types of measurement on questionnaires;
1.) Open ended questions
2.) Multiple choice questions
3.) Thurston scale
4.) Likert- type scale
5.) Rank order questions
6.) Check list
7.) Semantic differential
For today's' question for thought, I would like you to create a research question, and discuss how one type of measurement would be helpful to answer the question. In addition, I would like you to describe how one other type of measurement would not be the best measurement technique.
1.) Open ended questions
2.) Multiple choice questions
3.) Thurston scale
4.) Likert- type scale
5.) Rank order questions
6.) Check list
7.) Semantic differential
For today's' question for thought, I would like you to create a research question, and discuss how one type of measurement would be helpful to answer the question. In addition, I would like you to describe how one other type of measurement would not be the best measurement technique.
posted by Dr. Jeffrey Bolt at 10:39 PM
13 Comments:
Research Question: What is the effect of listening to music while studying on grades?
One type of measurement that would be helpful to answer this question is an Open Ended Question. I say this because since people favor different types of music, an open ended question will give you the most honest answer from each surveyer.
A Forced Answer/Multiple Choice Question would not be most appropriate here because then the surveyers may feel forced to agree that listening to music while studying is beneficial to his/her grades and also the choices may be set up in a way in which the surveyer would have to pick a genre of music to listen to and their desired choice may not be an option. Therefore, they end up picking one just to pick one which results in a dishonest answer.
Krysten Reed #2
I'll go alittle broad with krysten's research question and say what is the effect of outside interference while studying on grades?
Another type of survey that you could measure this is in a Rank Order Question. The reason for this is because an outside interfence can be your cellphone ringing, to music playing, and even your best friend talking to you. This certain measurement would make the person order in their own mind the worst outside interfence to the best or what doesn't bother them. You can measure individual frequency and since each and everyone of us is unique and different in our own way, it would be interesting to see if one outside interference bothers us all.
A check list in my mind would not be the best bet here. A surveyer will probably check all the possible answers and they would just be ranked all the same. The research wouldn't get to a clear point and figure out which outside interference is the worst. Some outside interference is alot worse than others. The check list would defeat the purpose of what students can handle and can't.
D Newman #2
Research Question: What is your favorite activity to do on the weekend?
One type of survey that could measure the responses is a check list survey. I choose this type of survey because it gives a list of activities that a person could do on the weekend. Also the participants could check all of the activities that are their favorite.
A forced multiple choice would not be the best survey to do in this situation. My reason for saying this is because there may be a number of activities that a certain person may do on the weekend. On a multiple choice survey you are forced to choose between answers, and you would not be getting an accurate description of what people’s favorite things are to do on the weekend. With the check list you can check everything that you do unlike a multiple choice.
Melissa Giancola #1
If one is seeking to determine how listening to music while studying effects grades, one might want to start by determining the types of music people listen to. As Krysten suggested, an open ended question, such as, (What kind of music do you listen to while you are studying?) would be a good starting point. However, there are inherent problems with this open ended question. For example, if the person does not listen to music while they study the question is invalid. Considering D. Newman’s suggestion of creating a question to notate rank order might be a good option. But here, too, one runs into the problem of getting to the meat of the information the researcher is seeking. Again, what if the subject does not listen to music? Can the researcher gather any good information from this subject? Creating the right question for a semantic differential scale could be a good measure of the effect of music on an individual while they study. A better choice might be a series of Likert-type scale questions such as:
When I listen to music while I study, I stay focused on my studies.
Strongly Agree 1
Agree 2
Neither Agree/Disagree 3
Disagree 4
Strongly Disagree 5
Mark N. section 2
Research question: What is your favorite class for your major?
I think a good type of measurement question for this would be rank order. You could put every comm class and have the student rank them in order from favorite to least favorite. This gives you straight forward answers.
A bad measurement question would be the liker type scale. People could have the same opinions about different classes. This would make it more difficult to calculate the most favorite class.
Brenna Teaberry #2
Research Question: What type of contraceptives do you use during sexual intercourse?
This question would only be given to people who are sexually active. I actually did a research project senior year of high school about teenagers’ use of contraceptives. I passed out multiple choice surveys with choices of all the possible contraceptives individually (including not using contraceptives), as well as the pill/ condom combination. I also had a place where they could add any comments. I found out that I did not give them all the contraceptives combos they used to check from. My research was then inaccurate because some people probably just picked one type instead of commenting that they used a combination that was not a choice. Therefore, multiple choice would not be a good measurement technique.
If I could do the surveys again I would turn it into the check list where people could check all the different kinds of contraceptives they most often use. I would then be able to accurately measure which type(s) of contraceptives teenagers are using and not using.
Deborah Pritchard #1
Research Question: how does violent video games affect the aggression levels of teens?
A type of question that would be helpful would be using a type of question based on the Thurston scale. The teens taking the questionnaire would choose statements he/she identify with most. Because the statements would be weighted, researchers could understand the From those statements, the researchers can collect data and measure levels of aggressiveness.
A type of question that wouldn't work would be rank order questions. If a participant had to rank a series of choices, for example ranking things occurring in everyday life that bother the participant, those choices marked by the participant would be hard to compare. Although the first and last choice would be easy to compare, every choice in-between wouldn't be as valid.
Earl Coburn #2
Research Question: On average how many times a week do couples have sex?
One way of finding out the results accurately would be to give an open ended question. By asking how many times per week couples have sex you would get an honest answer that will probably vary from couple to couple. Of course this survey would only be given to people who are in "relationships."
A multiple choice question would not be appropriate here because it would be forced. By this, I mean there would be a certain about of answers to choose from and those particular numbers may not be the amount that you and your "partner" engage in intercourse. Therefore, they would be choosing a number that would be inaccurate in the poll and then the whole poll would be inaccurate.
:)
LF #2
(T/TH-3:45-5)
Research question: What hero/role model do you look up to the most?
By using the rank order system this question could be answered most effectively. Options would be given such as: mother, father, grandparent, celebrity, friend and then the person taking the survey would rank from highest to lowest who they looked up to the most.
Using the likert-type scale would not be effective because there is no ranking of the hero the person prefers. A persons mom and a celebrity could both fall into the same category (such as agree) making it difficult to understand weither it was the mom or the celebrity that impacted the person most. If the question asked allowed for more than one hero to be ranked highest then this scale would work.
K.C. 2
Research Question: On average how many hours of sleep do you get a night?
One type of measurement that could be used would be open eneded question because it varies from person to person how much they sleep so it would be easier for them to just write the answer themselves instead of choosing an answer. We would probably get more accurate results. If we used forced ansewr their selection might not be there and they might chose whatever was the closet to their choice and we might not get correct results.
Brittany Kozak #2
Research Question: Do upperclassmen drink more than lowerclassmen?
I would use the Likert Scale and say that lowerclassmen drink more than upperclassmen. SA A N D SD and then I would conduct a survey.
Another type of measurement that may be helpful is the multiple choice question. I would ask questions surrounded by the number of drinks the different classmen consume and how often.
Marie Payden #2
With a question like "what did you have for breakfast?" it would be appropriate to leave it open ended, or a checklist as there are many possible answers. Thurston or Likert scales would be inappropriate since we aren't asking about how frequently they do something or whether they agree with statements. Multiple choice would also be inappropriate due to the inability to calculate a mean between breakfast foods.
Jacqui Jezek (1)
Do long distant relationships work out less than relationships close in proximity?
I think that close ended questions would be helpful in studying these two relationships, pintpointing certain areas of each relationship. Also, likert type scale questions and semantic differential scale questions could be used during this study. I don't really think that a check list would be helpful because I am not sure what could be assessed in this type. Also, I am not sure if rank order would be helpful either.
Heather Sichi, 1
Post a Comment
<< Home