Introduction Getting into a college is difficult, but studies now prove that staying in that one college, is just as difficult. Often times, students believe that college would be just like high school, and that they won’t have to study as much. However, to keep up with the grades, those college students soon discover, that they have to give two times their regular effort to do well. As years go by, college retention and graduation rates seem to be slipping for some reason. Reasons for this is because, of how your parents might or might have not went to college. If your parents went to college, then you have a 60% chance of graduating a four-year college, whilst if your parents didn’t go to college, you have a 39% chance of graduating on time. There is a big difference between 60% and 39%, and it is hard to believe how because of your parent’s educational past, that might take a toll on your future. Not only that, but where your parents stand financially, might also be a reason on college retention and graduation rates. Usually, to obtain a bachelor’s degree, you have to be in college for four years, so if you start off college at 18 years old, you’ll graduate by the age of 22 years old. However, studies show how 90% of high income born freshman are able to obtain their degree in four years, while approximately a quarter of the freshman who are born in the bottom of income distribution obtain a degree in six years, when they are 24 years old. Due to things that we are born into, such as our parents education, and their financial stability, college students might suffer, and this is why we have many risks in college retention and graduation. The SATs are a test that determine your eligibility in a college, and soon predict whether or not that college will accept you. Having a high SAT score means a higher chance of you getting into that college. However, SAT scores do not predict your eligibility of staying in and graduating on time at college once you are accepted. It is proven that those students who come from financial stability tend to do better than those who are poorer off than them, which is why a rich student with low SATs scores tend to graduate faster rather than a poor student with high SAT scores. Now universities address college failure and drop-out of at risk students very thoroughly, as they want their college to be standing well academically. During 1999 at the University of Texas, a professor of Chemistry named David Laude decided to intervene so that the at-risk college students would be able to graduate on time. So, within the beginning of the fall semester, Laude went through the files of each and every one of his chemistry students, and had found 50 students who he thought could be indicators in determining failure to graduation on time, or doing not so well in the class. These 50 students had low SAT scores, a low family income level, and parents who weren’t as educated as the kids had planned to be. Eventually, these 50 students were invited by Lauder to participate in a new program, entitle Texas Interdisciplinary Plan, where it would be a smaller section of Laude’s chemistry class and the students in this program would start to learn the same material that they might see as a bit more difficult, but what the kids in the larger section of his class were studying. So, Laude went from teaching at his 500-student chemistry class, to teaching at his 50-student chemistry class, with the same material, lectures, and tests given to the students, however the only difference was the actual class size, and who was in what class. After conducting this, the students in his 50-student class were getting the same test grades as those in his 500-student class, and in their fourth year, had graduated at the University of Texas, higher than in years passed. According to the study that I am designing, I hypothesize that the students who receive the Supplemental Instructions, or SI will see improvements in their grades, while the students who don’t take SI will continue to not do well and see their grade either staying the same, or even decreasing overtime. Method (i) This experiment should be done solely on the at-risk students in Dr. Steph’s class for introductory psychology. This means that anyone who is doing poorly on class tests, which make up the majority of grades in college, should be studied upon. The target population amongst all the students would have to be the students that are actually failing the course with a D or an F, because these are the students who need the most help in passing the tests, to graduate from the class in passing terms. The participants are selected and recruited for the study by looking up the grades of all the students in Dr. Steph’s classes, and finding all the students who are getting either a D or an F in the class so far. The sample amongst the target population would be random so there can be more accurate data, and so that there won’t be any unfairness in this study. (ii) Getting students to understand and actually paying attention to something is very difficult. This is why there should be constant reminders to address Dr. Steph’s concerns about the reluctance of at-risk students, such as writing it in the front board before class begins where all the students must look to pay attention to the lesson, handing out flyers or cards that are colorful and fun to look at, to get the students excited about actually going to the SI sessions, constantly pressurizing the students to do well by gently scaring them and telling them how it is especially so important to pass your classes in college to do well in the future, by sending out emails, and even maybe a note to serve as a reminder under every online assignment. In this, I think the smart method would be to test out four different groups. The first group will consist of the at-risk students who go to SI, next group will be the at-risk students who do not go to SI, the third group will be the students that are not even at-risk but still attend SI, and the last group will be the other students with the high grades that do not attend SI. The groups will have 30 students (because that is how many a regular class usually has) and will have 120 students apart of this study overall. In order for there not to be any bias, or for the study to seem as though there are flaws, there should be a random assignment of all the students into each group, (however this depends on their academic standing first). (i) There are two at-risk failing groups in this experiment, and of those two groups are those who will take SI and those who will not take SI. For those who take the SI class, there will be a psychology graduate who will help and teach the class, so that you can see whether the SI is actually helping the student or not. Now, some students are shy and don’t like raising their hands and answering questions. So, in order to help with that, have each student rip out a sheet of loose leaf, and write down any questions they have regarding the subject, so that in the SI sessions, the graduate teacher could help. The lessons should have colorful pictures on it, as students seem to pay attention more to color, and there could be videos being used that not only help the viewer understand the topic more, but it could be funny, so that the students couldn’t forget about them, and so it could help them on tests. The controlled participants would not be attending the SI session; therefore, they wouldn’t be receiving the extra help. Because this might be a little unfair to the students, have announcements in the beginning of the class, in which shows the students that whether they are in the SI group or not, they could still email Dr. Steph and ask for any help or questions they might have had within the lesson. Dr. Steph could also send the videos and slides that she shows in class to all her students, so just in case they weren’t listening in class, or they forgot some of the information that they were taught, then they could go home and revive all the information. In order to calm Dr. Steph about the control group in not getting help from the SI groups, you can show her more ways to provide the help that the students need in their lessons, such as having each student write down a ‘complaint’ or something they need improvements within the class or lessons, or questions that they might even have during the lesson, and then having Dr. Steph trying to make the class more beneficial for the students. (ii) (1) The problem with the controlled participants is that there could be no change amongst it throughout the whole experiment. The only way that students might mess this up would be by going to their TA’s or teaching assistants. In this case, the teaching assistants would be acting similarly to those teaching the SI sessions, so then they would be like an experimental group as well. In order to prevent this, TA’s should be taken out after class, the only TA’s that are available are the ones in Dr. Steph’s class, that everyone is able to question equally, and to be honest, in class, students are more likely to pay attention to their professor rather than the TA, and the TA would not be available after class. (2) For those students that have work outside of class and must rush to their work right after class, they would be problematic to the study. Since this study is done at random depending on your academic standing, you won’t know which of the students has work ahead of time. So, the only thing to do would be to ask the students to write down the questions they had by the end of Dr. Steph’s lecture, and the help that they needed, and to email whoever is teaching at the SI session, and a wait for the response, or by having the entire SI session videotaped and then sent to whoever couldn’t attend because of having to work, and then the students watching during their free time. (3) In order for there not to be an inconsistency in the SI sessions, make going to the SI sessions enjoyable for the students, so enjoyable that they’ll never want to stop going for the extra help. This can be included in the teaching methods by having a teacher that doesn’t bore the students, and instead shows them educational yet funny videos, and teaching them songs to remember some things that they learnt about. In order to get the students to continuously come to the class, you can give every student a stamped card for every time they go, and every stamp would be equivalent to test points. In this study you would need to control the amount of work that every student gets, so whether or not the student goes to SI, they should always get the same amount and level of work, and they should only decide to do any extra work at home if they decide to on their own, so the SI teacher won’t be able to provide them with extra work to do in their free time, that’s their own decision to make. In order to make sure that other than the SI sessions, all students are treated equally, you can have Dr. Steph email the students the same extra help in worksheets that could be beneficial for test prep, or by having an open emailing to the professor where each and every one of the students are able to ask for any help regarding the topic, or class. The dependent variable is the test scores of all the students. One effect of going to the SI sessions could be improved test scores over time. Another effect could be of students understanding the subject more, therefore having discussions in Dr. Steph’s class. The third effect could be of an increase in the number of students who decide to stay and decide to graduate in the psych major. The independent variable would be the SI sessions. The controls of the experiment should include asking help from Dr. Steph, having Dr. Steph email each and every one of her students the same slides, worksheets, and videos that could help for test prep, and being available to each and every one of her students when they need help in that topic. Results You would perform a statistical test on the experiment that you had designed for Dr. Steph by having each group take three psych tests in class, each after two weeks. You would calculate the differences within the groups, by finding the individual grades of each student, then averaging it up, and then gathering all the averages of the 30 students, and averaging up that data. You then compare the data amongst each of the four groups. You could do a paired t-test as this would be a comparison of the before and after testing of the groups on their exams. Such as the before being the test scores that were resulted before SI sessions began and the test scores that were resulted after the SI sessions began. This would be done by also using the means of the whole group compared to one another. The variation within the group might be with how many questions were asked by the students in the SI sessions or the likelihood of even attending each of the SI session classes by each of the students. Statistical significance is the probability of chance occurring during an experiment. Finding statistical significance within an experiment is difficult, as this is not done because you might think that the experiment, or the independent variable is what is resulting in the change, but it’s actually the statistical significance. You might have a statistical significance if the resulting data is either much higher or much lower than the rest of the experimental group. Conclusions No, Dr. Steph would not be correct in concluding that after undergoing SI, any at-risk student is at least capable of passing her course. For starters, the two students that are at-risk, come for help every day after class. They know that they aren’t doing well, but instead of just going home and being sad about it, they take time out of their own day to stay and ask questions about topics that they might need help in. These students show a passion to pass the class, that not all at-risk students may have. With this experiment, the experimental and control groups are with two students each, which is too small, compared to the actual group of all the students. The at-risk students are driven to do well, (this is proven by how they ask for help on the daily) that any type of assistance would help them do well in the class, so if they went to the SI sessions with their determinations, even with a little help, they would be able to do a lot better. The pilot students wouldn’t be receiving any help, they could only help themselves, which is why the at-risk students did far better than the pilot students in this mini experiment. If SI had nothing to do with the differences amongst the pilot study, then some differences can include: The at-risk students who continuously ask for help from the Dr. Steph after class. They might be able to make study groups with each other to help them in the studies. They might make study guides, and study from them every day of every week before a test, so they can smoothly memorize and understand everything in a clearer manner. They can also make practice tests for themselves on a daily basis to test them on that day’s lesson, so eventually they could do well on tests. This could all be a reason why the at-risk students received a higher grade than the control students on the exam. Dr. Steph would be able to improve her study if she has more students participating in the study, and if she spent a longer time conducting the research, resulting in more accurate data.