the literature is filled with studies aimed at identifying the causes of attrition and strategies to increase retention. the journal of college student retention (http://www.collegeways.com) is an excellent source of information on the subject. the statistics are discouraging: despite the tremendous expenditure of funds and manpower, retention has not improved over the last two decades. the lack of success is not because colleges and universities haven't worked to design programs to ease student transition into the academic and social systems. rather, these efforts have failed because of an inability to quickly identify the students 'at risk' and deal with the issues they face. each year just under 3 million students graduate from high school; only about 1.8 million of them will attend college, and fewer than 1 million will graduate.

seidman ["retention revisited: r=e,id+e&in,iv," In college and university, 71, (4), 18-20, 1996] reports that six year graduation rates for full-time students at four year institutions dropped from 52.6% during the period from 1977-1983 to 46.7% from 1986-1992. seidman, in his literature review, suggests a clear formula for cutting attrition:

retention = early identification followed by early intensive intervention

early identification is critical. if you don't learn that a student is struggling in english or mathematics until mid-term, it is unlikely you can intervene successfully. but if you can identify students 'at risk' immediately using sophisticated software that takes into account a rich array of variables, then you can design a program that will permit success. attrition results from a variety of factors and is not limited to students' struggling academically. retention expert may determine that out-of-state students in performing arts leave at an unusually high rate. that information provides you with the basis for developing appropriate strategies. for instance, you may need to schedule performances in a particular area to increase the sense of connection.

predictive edge's retention expert provides you with the information you need--early identification of students 'at risk.' the analysis of the data provides you with: early identification of the factors associated with high attrition, a tool to assess the performance of retention programs, year-to-year comparisons of program performance; it even lets you build retention predictions into your admissions decisions.

understanding retention expert

retention expert offers a number of valuable components, which taken in total, can form the basis of your enrollment management strategy. these elements include:

retention expert is as much a process as it is a product. we start by analyzing the data available from your student record database using data mining software.

data mining

through the use of data mining, we identify the principal variables associated with student action. for example, in a recent analysis of data from an independent institution, we determined that the strongest correlations (significant at 0.05 level) between the institution's data and student actions were:

data mining of institutional data and geodemographic data requires a number of intermediate steps to get meaningful data, but the outcomes provide important insights into student behavior, as well as the elements that will then be examined for inclusion in the model.

predictive indices

the primary premise underlying the use of all predictive indices, including retention expert, is that the traits associated with unsuccessful students are relatively stable year to year. this year's freshmen will behave, in many ways, like previous years' freshmen. we use this information to build an expert system that identifies both the student's 'at risk' and the factors that contribute to high attrition. predictive indices are constructed in two distinct forms--static and dynamic. a static index depends on parameters assembled early in the process; these parameters typically do not change during the process. a static index might include the high school, geodemographic data, act or sat scores, high school gpa, high school activities, and any other relevant admissions data. by contrast, the dynamic index portion of the index is updated throughout the process as new information is logged; this would include college gpa, college activities, leadership roles, etc.

this construction then allows for differing uses of the predictor throughout the academic cycle. the static or 'fit' index can be utilized to make strategic decisions early in the admissions cycle. the fit index ensures that the academic and student affairs staffs are directing their efforts to students in need of assistance. the dynamic or 'hot' index provides a continually updated measure of the probability of attrition as the academic year progresses.

components of a predictor

a predictor uses data from the recent past to anticipate the near future. the choices made in establishing the database are critical. if the model is constructed using only one year's data, there is the danger that a particular year was not typical. if the model is based on multiple years, there is the danger that conditions have altered significantly over the time period. a critical assumption for any predictive model is that the boundary conditions are unchanged: an improved economy might positively impact college retention; a major campus crime might have negative impact. weighted averages and restricted time spans help avoid these dangers.

a single variable predictor is the simplest to calculate. the major of the student, for instance, represents a single variable that will be associated with attrition. multiple variable models are richer and become increasingly predictive as parameters are added. add the sex of the student, test scores, high school rank, etc., and we can refine the likelihood of that student's remaining at the college. an index is then a mathematical combination of a number of probabilities, based on past data, that predicts the likelihood of a future event.

proving it works

predictive edge uses five years of prior student record data with known outcomes to develop the model. we then test the model on the most recent year's data to determine its effectiveness. for instance, the data from the fall 1996 to fall 2000 cycles would be used to establish the model. the data for the fall 2001 would be used to test the efficacy of the model, and the algorithm would be used for the fall 2002. this approach provides a robust model that can be used effectively for two or three years.

outcomes - examples

the modeling algorithms developed by predictive edge have been applied to data from a number of different colleges. these institutions differed in location, mission, and size. in all cases, a unique algorithm suited to that institution was developed. the table below shows typical output for enrolled students.

saving money with retention expert

there are many advantages to using retention expert. it will:

although listed last,retention expert's capacity to identify students with virtually no prospects of succeeding is one of its most powerful features. (retention expert does not use ethnicity, religion, etc., in its models.) most colleges tend to admit students in the same pattern year-to-year. the result is that they continue to commit the same errors over and over again. the college then expends extraordinary energy and funds trying to assure the success of students with little probability of success. the waste can be from $50,000 to $200,000. it also discourages faculty and counselors, not to mention the negative impact on the mismatched students. at the same time, students with a far better chance of success are ignored and ultimately leave.

pricing for retention expert

retention expert includes:

retention expert will pay for itself many times over. because of the robust nature of the model, it normally retains its effectiveness for two or three years. for a modest fee, predictive edge will check the efficacy of the model each year, provide year-to-year comparisons, and inform you when a new algorithm needs to be developed.

the price of retention expert is based on the breadth and scope of the work. Contact us today to discuss retention expert and receive a quote.

unique advantages of retention expert