Class Sizes – Better Summary Measures than USNWR?

<p>I don't buy itnto that research. Many people learn best on their own. Others just slow me down. Many people learn better by sharing too, and in their case, I would imagine that a smaller class would be optimal.</p>

<p>i-dad:</p>

<p>it seems intuitive that smaller is better than bigger. However, the data for K12 is sparse; essentially only one, small study but it wasn't really double blind, so not sure many conclusions can be drawn from it (a small point that educators ignore).</p>

<p>Do you know of college-level studies? (I just have never looked for them)</p>

<p>
[quote]
Many people learn best on their own.

[/quote]
</p>

<p>In that case, we should do away with professors and classrooms altogether and just go to online lesson plans in a PowerPoint like format.</p>

<p>We also need to consider that there are several different kinds of "learning". One is simple learning of facts as measured by a standardized achievement test. That is very different from learning how to formulate, argue, and defend a theory both orally and in written form. That kind of learning is difficult without opportunities to do it and get feedback.</p>

<p>"In that case, we should do away with professors and classrooms altogether and just go to online lesson plans in a PowerPoint like format."</p>

<p>Professors at top universities teach material that has not hit textbooks yet. Many students, myself included, did not go to college to learn intro level material. I went to college to learn groundbreaking, graduate-level material. </p>

<p>And learning to formulate, argue and defend theory is ALL education should be about. Some do it best in groups, others do it better on their own. You must acknowledge that there are different learning styles and that many people do not learn as effectively in a group setting. </p>

<p>At the end of the day, it is about personal preference. From my extensive experience on this subject, people learn</p>

<p>bluebayou:</p>

<p>Here's a short article on teaching college writing that will point to much of the available research:</p>

<p>Class</a> Size in College Writing Classes</p>

<p>It's not at all conclusive because it depends how you measure effective teaching. Much of the research uses standardized achievement test scores as the outcome measure of learning as absorbing facts. That, of course, is a different experience than teaching someone to write essays.</p>

<p>I'm curious about the Kent State research showing that students in smaller freshman writing courses got better grades in their courses the following year.</p>

<p>articles on class size by Cornell researchers</p>

<p>eCommons@Cornell:</a> The Effects of Class Size on Student Grades at a Public University</p>

<p>The</a> Class Size Controversy</p>

<p><a href="http://www.arts.cornell.edu/econ/seminars/Urquiola.pdf%5B/url%5D"&gt;http://www.arts.cornell.edu/econ/seminars/Urquiola.pdf&lt;/a&gt;&lt;/p>

<p>The</a> Effects of Class Size on Student Grades at a Public University</p>

<p>The</a> Effects of Class Size on Student Achievement in Higher Education: Applying an Earnings Function</p>

<p>It is interesting to see the enrollment caps on courses at large university X. It is almost like they occur in increments of 5 and 10. At this university, 95% of their classes have fewer than 40 students.</p>

<p>How much does it correspond to the physical size of classrooms?</p>

<p>enrollment cap, number of courses</p>

<pre><code> 1 14

2 8

3 2

4 8

5 57

6 8

7 5

8 24

9 7

10 144

11 1

12 41

13 12

14 18

15 244

16 44

17 7

18 30

19 27

20 301

21 4

22 20

23 12

24 29

25 178

26 4

27 108

28 28

29 4

30 185

31 4

32 129

33 32

34 7

35 243

36 47

37 1

38 20

39 2

40 180

41 2

42 14

45 24

48 7

50 16

53 1
</code></pre>

<p>collegehelp,
I don't know what the school is, but if they were trying to game their statistics, wouldn't they be better served by capping the numbers at 9, 19, 29, 39, 49, 99?</p>

<p>hawkette-
That's probably true. I guess they're not on top of things. haha</p>

<p>Most of the higher education-geared papers that Collegehelp provides from Cornell are pretty clearly based off of data from SUNY-Binghamton.</p>

<p>
[quote]
collegehelp,
I don't know what the school is, but if they were trying to game their statistics, wouldn't they be better served by capping the numbers at 9, 19, 29, 39, 49, 99?

[/quote]
</p>

<p>Hawkette, isn't it refreshing to find a college that isn't trying to game its statistics? If 21 seats can fit in a classroom at firecode, than why not educate 21 students? Do you honestly think students would learn better in a 19 person class than a 21 person class?</p>

<p>I can only speak for my own alma mater, but at Cornell the emphasis was always on providing the best classroom experience to the most number of students within the scope of resource constraints. </p>

<p>So if that meant educating 32 students instead of 19, that's what they did. And if that meant educating 1000 students in Psych 101 instead of 150, that's what they did.</p>

<p>At the same school, there isn't much difference in average lecture class sizes among disciplines. They all hover in the 20-25 range. Math courses, physics, and the various social sciences are the exception and tend to be large while the various art/music disciplines tend to be smaller.</p>

<p>cayuga,
I'm with you on this issue and I appreciate the institutional integrity. I read charges often on CC of schools manipulating their numbers. I don't know how accurate these statements are and frequently I think that they are overblown and mostly based on squat…and maybe even jealousy. </p>

<p>As for class size measurements, I favor smaller over larger, but agree that a difference of 19 vs 21 is not meaningful. It's just that they have to draw the lines somewhere and the CDS measurements were set at 2-9, 10-19, 20-29, etc. Would you do it differently?</p>

<p>Are academic standards compromised in small classes?</p>

<p>Faculty get to know students and don't have the heart to give bad grades?
Methods of evaluation are "looser", more subjective?</p>

<p>Well, I haven't done much reading or research into this topic, but I think for college level courses you can basically break down the types of experiences into four rough divisions. </p>

<p>I'll also note that this divisions are really only for the high end of the higher education spectrum... we're not really talking about a large state school or community college here, even though their challenges are probably much more deserving of our time.</p>

<p>1) Tutorials. One or two students and a professor. I had the privilege to study at Oxford, which has a tutorial system, and to have three tutorials/independent studies while at Cornell, and it's a fantastic learning experience for upper-level work. I'm not convinced of its benefit for introductory coursework, however.</p>

<p>2) Courses that require small group discussion and seminars. Most humanities courses can benefit from this, as well as more "exploratory" (e.g. non-lecture) based) courses in the social and natural sciences. Most people agree that these classes should be around 20 students or less, but if it is capped at 15 or 21, it probably doesn't matter. I can see how a seminar of only 7-10 students may actually result in not enough views or contrasting opinions being brought to light.</p>

<p>3) Lectures courses where any student can feel comfortable asking a question at any time, and the professor will generally know the student by the end of the semester. I generally think that these lectures feature 25 - 60 students. At major research universities, a lot of introductory and intermediate social science courses can fit in this mold, where your professor will be prolific in the field and there will be a lot of student demand to "consume" these professor's lectures. </p>

<p>One of my advisers at Cornell was Ron Ehrenberg, who is a pretty widely respected expert on the economics of higher education. His courses typically featured 40-50 students, every single one of them sat captivated through lectures, and he was on a first name basis with most students by the end of the class. So we didn't care that the class size wasn't under 20 students.</p>

<p>This is really where teaching quality matters, and where the LACs will tell you that the smaller class sizes are better and the national universities will tell you that having experts in the field are better. Personal preference plays a large role here. Do you want your hand held or do you want your professor to have testified in front of Congress?</p>

<p>4) Large introductory courses where you are treated like a number. Think 75 or more students. This happens at all major research universities, as well as a fair number of liberal arts colleges, and what's most important here is whether or not the course is structured well, and that open office hours and TAs are available for questions. These lecture experiences are most likely to occur in the natural and life sciences, as well as introductory social science courses. I can only speak for Cornell, but they tended to do a phenomenal job of running these types of courses. Nobody complains at Cornell about Psych 101 having over 1,000 students; it's really more of an asset.</p>

<p>So I suppose the less than 20 students and greater than 50 students breakdown holds, but the major problem with the data is that it is omitting all independent studies and honors thesis tutorials, which is a pretty glaring omission of the type of faculty contact a student can expect to have. The other problem with the numbers is that they fail to control for discipline. Certainly a school like Cornell, which is heavily tilted towards the natural and life sciences (engineering, biology, etc.), might be expected to have larger class sizes simply because those disciplines lend themselves more to lectures than small group discussion seminars.</p>

<p>If you compare the numbers for Princeton and Cornell, and assume 5 classes a semester, you will see that whereas a Princeton student graduating with 40 classes can expect to have 14 classes with less than 20 students in it, a Cornell student will have 10 classes. And a Princeton student will have 18 classes with over 50 students, whereas a Cornell student will have 22. So the net benefit of Princeton, at least in terms of class size, is 4 more seminar-style courses over the course of an academic career.</p>

<p>That's not a trivial amount, but it's not that big of a difference either, especially for everybody who spreads the fiction that at Cornell "you will always be treated like a number". Especially in light of the fact that Princeton's endowment is worth about six times that of Cornell's on a per student basis, one really has to wonder where all of Princeton's money is going.</p>

<p>I agree that the charges of "gaming the system" are probably pretty farfetched. Course size is determined by a lot of things (resources, space, student demand, faculty input) and the people with the authority to manage those factors aren't necessarily the people who are filling out the surveys and sending data to IPEDS and the like. Colleges and universities have a lot on their plate--adjusting their courses solely to alter a single statistic that may or not be looked at by prospective students....it just doesn't make good sense .</p>

<p>Thinking more about class size...</p>

<p>If a school offers more than one section of a course, that to me indicates either:
(1) an attempt to limit class size for that course
(2) a desire to make the course available at more than one time of day
(3) a desire to offer somewhat different versions of the same course</p>

<p>If there is only one section of the course offered, then the class size is probably based on student interest. Put everybody who wants to take the course into one classroom.</p>

<p>At large University X, here are the percentages of classes in each size category that had more than one section, perhaps indicating an effort to limit class size for that subject:
2-9, 44%
10-19, 58%
20-29, 68%
30-39, 73%
40-49, 81%
50-99, 70%
100+, 44%</p>

<p>What does this say to you?</p>

<p>56% of the classes in the 2-9 category had only one section to accomodate all the interested students. In the 10-19 category, one section was enough to accomodate all interested students for 42% of the subjects.</p>

<p>This is a good point, collegehelp. At Cornell the intermediate economics courses were limited to around 50-60 students (depending on the size of the lecture hall) and there were typically three separate sections of the courses offered per semester to satisfy student demand. </p>

<p>So the idea here is that the Cornell economics department found it educationally worthwhile to split a 150-180 student course up into smaller sections of 50-60 students. However, they did not find it beneficial to create class sizes below 40 or 20.</p>

<p>Referring back to post #66, for the common data set categories 2-9, 10-19, 20-29, and 30-39 the actual average enrollment size was very close to the midpoint of the range. For categories 40-49 and 50-99, the actual average enrollment was closer to the low end of the range. For 100+, the average enrollment was 171 based on about 15 courses.</p>

<p>This might be of interest to the OP.</p>