ReplyA minimum of 100 words each and References Response (#1 – 6) KEEP RESPONSE WITH ANSWER EACH ANSWER NEED TO HAVE A SCHOLARY SOURCE with a Hyperlink
Make sure the Responses includes the Following: (a) an understanding of the weekly content as supported by a scholarly resource, (b) the provision of a probing question. (c) stay on topic
1. Great post! When a researcher has each subject participating in more than one condition of the experiment, this is considered within-subject design. This could become tiresome if there are many conditions that are drawn out. When the subjects participate in more than one condition, it increases the chances that a researcher can identify differences between the conditions. This increases the power of the experiment from a statistical perspective. Since the subjects are measured on the dependent variable after each treatment, this design can also be called a repeated-measure design (Myers & Hansen, 2012).
2. Within subjects is when you have one group of participants that are exposed to both experimental conditions (Myer, & Hansen, 2012). For example: you have 10 participants (participant # 1-10), and two conditions for the experiment. (condition 1, and 2), participants 1-10 are exposed to condition 1 first, and followed by condition 2. So in condition 1 you have participants 1-10, and condition 2 you have participants 1-10 as well. Meaning you get twice the data from one group. Between subjects you take two groups of 10 participants (participant #1-20) each, and you expose each group to a condition, so condition 1 you have participants 1-10, and condition two participants 11-20. You would use within-subject design when the conditions in the experiment are not affected by each other, or when the first condition does not give away the purpose of the experiment. Small N design (one or a small amount of participants) is used for case studies or unique conditions. Large N design (large amount of participants) looks at the more general results of a population to get a more accurate result (Myer & Hansen, 2012).
3. When a study is needing to compare multiple interfaces, a between-subjects design can be used or a within-subjects design. It will simply depend on if the study will use the same participants or different participants. Between-subjects is used when different people test each condition so that each person is exposed to only one user interface. Within-subjects is used when the same person tests all the conditions. An advantage is that you can use a smaller sample size. Studies spend a lot of money on recruitment, honorariums and facilitator time, therefore, within-subject studies is good at reducing such costs.
Small-N designs involves observations of one person or small groups, during and after an intervention period. Large-N designs look for patterns in a large number of cases. Large-N studies have better external validity, A small-N approach has better internal validity. Sometimes the independent variable will dictate which experimental design to use for a between-subjects design or within-subjects design. Often times, both designs can be used. A large N design can be used when there are enough subjects and researcher want to increase generalizability.
4. I enjoyed the information in your post. As you mentioned counterbalancing is a great way to minimize fatigue, boredom and practice effects. In the place of randomly assigning the participants to conditions, it would help to randomly assign the participants to different orders of conditions. Some type of random assigning must be present during an experiment. For example, one group can do the more exciting condition of the experiment first while the other group does the boring condition first. Then the participants can swap conditions. This will help because the difference in the dependent variable between the two conditions cannot have been caused because of the order of the conditions (Open Textbooks, 2015). I think when using the elderly, the researcher must be careful seeking elderly people who cannot withstand the different conditions. Some can endure longer than others. Someone who has practiced and studied the experiment long enough can affect the results if they figure out the experiment.
5. Thank you for sharing you insight to the DQ! I agree that the experimenter must add to the treatment condition some enthusiastic events to make the subjects enjoy the experiment. i would like to add that changes can occur when the subjects are in more than one condition; fatigue effects cause performance to decline as the experiment continues and the subject becomes tired, make mistakes, become bored and irritated by the experiment (Myers & Hansen, 2012), but then the factors may lead to improvement as the experiment proceeds of which is when practice effects take place and the subject is familiar with the experiment and does better with the experiment. Within the experiment both positive and negative responses are called progressive error (Myers & Hansen, 2012), when the experiment progresses the results are biased, hence the subjects’ responses are not caused by independent variables but by order effects produced by going through more than one treatment condition making for effects order such as effects of practice. I also believe that to minimize the problem of fatigue, boredom, or practice effects in a within-subjects designs is through varying an extraneous variable order, in this the experimenter can create a confounding counterbalancing (constantly letting the subjects experience a new treatment/condition without recognition), thus making for the subject to rate a new experiment higher than an old experiment, hence treatment condition would change because of the subject’s satisfaction responses (Myers & Hansen, 2012).
6. A within-subjects design can create many factors that will impact the results of the experiment. Fatigue effects are exactly as they sound, the subjects become tired (Myers & Hansen, 2012). Practice effects are created by the subjects becoming comfortable with the factors of the experiment and their results become more practiced and more accurate (Myers & Hansen, 2012). Boredom can cause the exact same response in subjects. To prevent this from happening we introduce counterbalances. Counterbalances distribute errors across the varying conditions of an experiment. This allows us to ensure that the order effects that may alter the results on one condition will be offset by the order effects impacting other conditions.