The purpose of this paper is to evaluate the differencesbetween at-risk students who utilizes SI and at-risk students who do not. Allof these students started on the same path, but this experiment reveals thatany student can succeed with a little bit of guidance and help. A) There are many risk factors thatlead students in college these days to a high risk of failing their courses ordropping them, but the two biggest risk factors in college learning andgraduation rate are the student’s family income and standardized test scores.

Studentswho have low income families tend to have many challenges in their daily lives.Many have money issues, which can overwhelm and cause stress that can lead to astudent to not achieve the needs for learning and getting a passing grade. Thetuition and loans can also overwhelm them, and force many to take up jobs whichmight be demanding and diminishing on the students’ performance in college.Another big risk factor in college students dropping out or failing, isstandardized test scores. Students that come from low income families tend toalso have low SAT scores, due to that fact that they don’t have the resourcesneeded to practice and prepare for the test. With lower SAT scores thesestudents seem to struggle more in college and seem to be the most at risk of failingor dropping out. 1B) SAT scoresdon’t predict college graduation, but instead the student’s family’s income iswhat determines if the student will graduate or not. Students that are borninto a rich family, tend to have a higher chance of graduating in 4 years, thancompared to students who are born into a poor or low income family.

Even thoughwhen comparing a poor student with high SAT score to a wealthy student with alow SAT score, academic success and graduation rate tends to favor the familieswho are rich. 1C) One interventioncolleges are now providing is extra classes and programs to better improve thegrades of students and understand those students who are at risk of failingclasses, so that they can help prevent them from dropping out or failing theircourses. As shown in “Who Gets to Graduate” by Paul Tough, one interventiontechnique they used at universities to address college failure and dropout forat-risk students was a new scholarship program. This program allowed at-riskstudents to improve their skills, and help students that have unmet financialneed, by giving these students who were in the program $5,000 every year. 1D) The hypothesis being addressed inthe study that I am designing is that at-risk students who are receivingsupplemental instructions, extra help, and extra classes will see an increasein their grades, while students that do not receive these benefits will have avery low chance of improving their grades and will continue to struggle.Thereare many ways to plan your research, but in all researches it is important tohave a consistent idea of what you are trying to learn from the research.

2AI)The participants would be the students from Dr. Steph introductorypsychology class because Dr. Steph is worried about the many students who areat-risk of failing her introductory psychology course at City College. The targetpopulation would be the students that are at-risk of failure, since Dr. Stephwants them to pass her course.

Since a passing grade in CCNY is considered froma range of C to A+, participants are selected and recruited for the study by gradesranging from F to C-, as students who receive these grades are consideredfailing. I would target the populationrandomly in order to make sure that there is no bias to this experiment and togive each student an equal opportunity. 2AII)I plan to address Dr. Steph’s concerns about the reluctance of at-risk studentsto participate by making announcements in class, through sending them emails,notifications on social media and notifications on blackboard. I will alsoinform them to join me after class for a short amount of time to talk about theSI and how it can benefit them and will also give them extra credit. To ensurethat at- risk students participate in going to SI, SI attendance will count asextra credit, and make it mandatory for students below a C grade. I will usethree groups for my research, two of which will be experimental and one ofwhich is the control group. In both experimental groups the students will haveSI but, in one experimental group the students have work (jobs) and in theother experimental group the students who do not work.

The control group willjust have students who do not get to have the SI sessions. I will randomlyselect about fifty at- risk students from each group. I will use randomassignment by letting a computer program randomly pick at-risks students bytheir student ID code for each test group. For example, if there are 200at-risk students in the experimental group with SI who don’t work, I willrandomly pick 50 out of 200. This will allow the experiment to be bias free andprovide a fair treatment to the students. 2BI) The SI intervention for the experimental group wouldinclude reviews of the lesson, homework help, test prep, essay rewrites andtutoring. For the control group there would be only independent studying andlimited online resources included.

To address Dr. Steph’s concern about intentionally withholdingtreatment from the participants in the control group, the students in thecontrol group will be provided limited online resources, which will featurearticles and videos. By doing so, the students will subconsciously believe thatthey are receiving the amount of knowledge to succeed the course, and thus notmaking them feel like they are being suppressed. 2BII)Thethree following variables, which are TA visits, outside work, and inconsistencyof attendance in SI sessions can be controlled by having the student make TAclasses mandatory and having a penalty for missing one class, which would be droppingthe student from the course overall. By doing so this would prioritize thestudents outside work and plan out strategically their workload. This willprobably cause them to be cautious and make a schedule of their dailyactivities, so it won’t interfere with the SI session. Also if the student has inconsistencyof attendance, for example if a student is late they would stay 20 minutes andreview what he/she missed and take a short quiz referring to the topic taught thatday. Two other variables that need to be controlled in the study is the amountof homework turned in each week and the amount of extra credit provided.

2C) I can operationally define thedependent variable by the grades the at-risk student receives in quizzes,homework and exams. Three possible beneficial measures that can happen to anat-risk student is that they will have a better understanding of the material. Anotherbenefit can be that the at-risk student gets higher scores on quizzes and examsbecause of SI. The last benefit can be that the at-risk student develops betterstudying and learning habits, such as involuntarily visiting their TA moreoften or start visiting if they had not before. 2D) The independent variables is whether the student will take SIor not.

The control variables are TA visits, outside work, inconsistency ofattendance in SI sessions and the reluctance of at-risk students.Whenwe find ourselves with the results of the research, we often stumble upon themeaning of it. So we turn to mathematics to reveal to us the magnitude of whatwe have discovered. 3A) A statisticaltest can be used to measure the test scores between the students who attended SIsessions and the students who didn’t. For example, we can find the differenceby calculating the range of quizzes, homework and test scores from the studentsat-risk of who attended and the students at-risk of who did not.

3B) The variation between groups can be calculatedthrough standard deviation. The first thing we need to do is to find the difference,which is calculated by taking each difference of all three groups, and then squaringthe average results. This has to be done twice since there are two experimentalgroups. 3C) Statistical significanceis the probability that the results of statistical test are not by chance.

Statistical significance can be either weak or strong, for instance if thestatistical significance is low, that the means it is least likely because of chance.I would conclude that the difference between groups is statisticallysignificant if the difference of the averages of the groups is bigger than two.If the at-risk students who attend SI sessions receive grades with B’s orhigher compared to students who did not attend the SI sessions, who stillreceive C’s and below would lead me to believe that thestatistical significance is strong.Atthe end the data has been collected the discoveries has been proved, the onlything left to do is to improve and learn from our findings. 4A) Dr.Steph’s conclusion is incorrect because the students who visited her were notout of a random sample from at-risk students. This clearly shows bias and also wedo not know for sure the other variables that might come into play, such asoutside work, class schedule, free time, and disabilities. 4B) It is possible that the at-risk pilot participants actuallyhave much better course performance than the comparison pilot participantsbecause there were only two volunteers compared to the size of the populationof the comparison group.

We also do not know what treatment the two groups got,perhaps the two students probably had more personal learning from the pilotstudy. Hence, this shows that this study does not represent the populationbecause of the lack of descent amount of participants and lack of variablecontrol. 4C) The SI had nothing todo with the difference between groups in the pilot study because the realsource of this difference could be that the pilot experimental group could havealready been doing better than the pilot control group, the pilot experimentalgroup could have received outside help/resources or the control group could nothave received any form of SI. 4D)I would recommend that Dr. Steph to gather more students through random samplefor her pilot study.

Overall Dr. Steph needs to provide a larger sample sizeand give both groups same level of treatment as much as possible. For instance,if you have a large experimental group, you should have a large control groupas well.