Hypothesis
IB Juniors given a list of 18 words with a list of 3 category titles will recall the items on the list more accurately than those given the list without the category titles
Null Hypothesis
There is no difference in recall between IB Juniors given a list with category titles and those given the list without the titles.
II. Methods: Design
Design
The design that will be used for this experiment is independent samples, because it meets the needs of the experiment and also minimizes error and is effective in controlling specific internal validity concerns. This specific design controls for testing and progressive errors, most order effects, maturation, and mortality. This design, however, does not control for history. The only unwanted subject variables in this experiment would be learning or memory deficit disorders. Drastic problems that keep a student from being able to memorize or learn new things effectively would negatively affect the experiment and make the results more difficult to generalize. To control for this, as a sort of “pre-screening”, only juniors enrolled in the International Baccalaureate program will be used as volunteers. It can be ensured that any student which has made it that far within that rigorous an academic program has no drastic disorders which effect his or her memory or ability to learn and retain information. This also controls for selection in that all IB Juniors are statistically equivalent. Diffusion of treatment in this experiment could be a problem if the subjects share words or information from the lists while trying to recall them. This could largely alter the results of the experiment and therefore the researchers will be preventing subject interaction and supervising them as they are recalling remembered words. All subjects will be given the exact same list of words, only ½ will be given simple category titles. The words will still be in the same order and place on the page to control for instrumentation errors.
Ethics
Because subjects will be tested in front of a teacher, it will be made fully aware that they will be ethically treated and that in no way will this experiment alter their status in that classroom. The researchers will establish a clear and fair agreement with the research participants prior to their participation. This agreement will clarify the aim of the study, that it is completely optional, and ensure that the results will be shared with the participants after all data is collected and analyzed. The researchers will make the subjects fully aware that the study is not only optional, but that the subject can decline to participate or withdraw from the research at any time without any repercussions whatsoever. All necessary aspects of the research will be described and the volunteers will give their informed consent prior to participation. Results will remain confidential and anonymous, but the final purpose of the study shall be revealed in a follow up session. Since all of the participants are under 18, the principal and the party who is legally responsible for the students while they are on campus (Mrs. Benware) will be made aware of the aim of the study and give her consent to let any volunteers participate in the study.
III. Methods: Participants
The entire population of IB Juniors may be used in the experiment; however, the statistical concepts of sampling make it unnecessary. The sample will consist of IB Juniors from Mrs. Jamison’s 3rd and 6th period classes, and from various other teachers which give informed consent for the researchers to talk to their classrooms about the experiment and offer the opportunity for the students to volunteer. The participants will be a representative group of the entire population of IB juniors. Students which are taking psychology don’t create any errors in population validity, as they haven’t participated in enough training in psychology for their results to be altered in any way because of it. To obtain a random sample from the IB Junior class, students from random classrooms will be asked to volunteer. A random sample of 30 IB Juniors is used in the experiment. They are all 16-17 years of age and have similar academic backgrounds and capabilities, as they are all in the International Baccalaureate program.
IV. Methods: Procedure
Materials
The sample experiment used two groups of materials. One half of the participants will be given a list of 18 words with 3 subject headings on the top which read, “The following words fall into three categories: Food, Clothing, and Furniture.” The other half of the
participants will be given the same list without the category headings. The subjects will then be given another paper for the recall part of the experiment. More than 30 copies of each paper will be needed, as the minimum number of subjects necessary due to the experiments level of measurement is 30. Any subjects tested past that number will need all papers to be properly tested and a subject testing with the other (category/non-category) paper to balance out the results. It is wise to bring writing utensils in case students aren’t equipped with them already, so a box of number 2 pencils will be brought with the researchers at the time of the experiment.
Procedure Used to Collect Data
On the day of the experiment, the researchers will go into predetermined classrooms and introduce themselves and inform potential participants of the nature of the experiment. The students will be informed and asked if they would like to volunteer in the study. Those students willing to volunteer will be randomly allocated into two groups of equal number. Half of the students will be given a list of 18 numbers with category titles, and half will be given an identical list without the titles. Those lists with category titles will read “The following words fall into three categories: Food, Clothing, and Furniture.” Aside from this one line, the lists will be identical. Both lists will be in the appendix.
The subjects will be told they have 45 seconds to look at the list of words. The papers will be collected and the subjects will be given a sheet of lined computer paper when the other papers are completely collected. Half of them will say “Category Recall”, and the other will be headed with “Recall” so the researchers will be able to differentiate between the two during data analysis. The subjects will then be given 1 minute to write down any and all of the words from the list they can remember, then they will be asked to turn over their papers for collection This experiment has been piloted by 12th grade students in the International Baccalaureate program and has been evaluated as appropriate and relevant.
Controls
Standardized instructions control for experimenter bias. Keeping the two groups separated and not allowing communication between any participants avoids diffusion of treatment. Running both groups in the same room, although separately, controls for constancy of condition. All other control issues were addressed in the design.
Results
The results of the experiment showed many ties and very similar data between the two groups. Overall it can be seen that the number of words recalled in total was more in group 2 than it was in group 1, which goes against the hypothesis. The median and mode of both groups were 10 and the mean varied only slightly. In fact, the group which was not given category titles had a larger mean than the group that was. The mode and median are significant in processing the data from this experiment because it shows how many ties there were in the data. The difference in the data wasn’t significant to produce a different mode or median.
Difference in Recall Between Students with Categories and Those Without
To get a more concrete analysis on whether the two sets of data showed any kind of significant difference between the two groups, inferential statistics were used. The Mann-Whitney U test was used to because the data met the requirements for an ordinal level of measurement and was taken from two independent groups containing a small number of subjects.
C(n1=7)
E(n2=7)
U= 16
Chart value of U=3
The value for U in our experiment was calculated to be 16 and the chart used to analyze this value said that U showed a significant difference with a 5% margin of error for our specific number of participants only when U was less than or equal to 3. Thus, there was no significant difference between the two separate sets of data and the null hypothesis must be accepted.
Discussion
Through statistical analysis of the collected data it has been shown that there is no difference in recall between IB Juniors given a list with category titles and those given the list without the titles. A brief glance at the means of the two sets of data may indicate a small difference, but the Mann-Whitney U test clearly illustrated the lack of significant difference between the two sets of data, although often this test is too crude to pick up slight differences within the data that may be of some significance.
The research contradicts recent theories about categorical clustering; however, these results are partially due to a low number of participants. The experiment provides no evidence against Weston Bousfield’s theories about automatic categorization because it didn’t offer any questioning to students not given the category headings over whether or not they recognized the pattern within the words on the list they were given.
A larger sample number such as 30 could have provided more accurate data and also been enough to run statistical tests that pick up many more slight differences than the Mann-Whitney U test.
All of the students were enrolled in the International Baccalaureate program and have been trained to be extremely adept in memorization and they have been trained to retain as much information as possible. The high level of academic ability was thought of as a good way to control for academic background, but a more mediocre level of academic capability may have shown the effect of the categories themselves instead of the effect of years of rigorous academic training
Many improvements could be made to this experiment to make it run more smoothly and perhaps produce results which are more significant. The major problem in this experiment was the lack of participants. Perhaps students of a younger age who haven’t been trained so much would give more relevant results to mnemonic studies. It has also been hypothesized that the time given to the students for them to write down the words was too long, and that if a shorter period was given the students would have to rely on impulse more than deep thought. This could be more effective in studying the direct effect of categorization.