enrollment in junior-college years is essentially local and the latter set of percentages bear out the statement in the report of the General Education Board for 1902-14 (p. 119) that "state lines have always counted heavily in determining the area of college . . . influences."

It is worth noting that the medians for the lowest and highest thirds of the colleges tend to approach each other as the distance from the local community increases.

Numbers Enrolled in Junior-College Years who are from the Local Community and from within 25 Miles.-Not percentages of students only, but gross numbers of students also in these first two college years from the local community and from within a radius of 25 miles will throw some light on the problem of the size of community in which it is likely to be feasible to install junior-college work. These sets of data are supplied in Columns 8 and 9, respectively. The former shows a strong tendency toward increase of the number from the local community as we proceed from the smaller to the larger municipalities. Some measure of this increase is provided in the medians at the foot of the column. The median for all the colleges is 51 students, while the medians of the lowest, middle, and highest thirds are, respectively, 37, 46, and 106 students. The tendency is not as notable for the numbers from within 25 miles, the medians of these three groups being respectively 70, 71 and 131 students.

The Number in Each Thousand of the Population of the Local Community Enrolled in Junior-College Years.-One additional relationship in the data here presented has meaning for the junior-college problem-the proportion of the population represented by those enrolled in freshman and sophomore college years. This has been computed in two ways (Columns 10 and 11 in the table). The first of these is the number in each 1,000 of the population of the municipality in which the college is located represented by those whose residence is in the municipality. This proportion is seen to be over forty per thousand for the colleges in the smallest com

munities and to tend to diminish rapidly as we proceed to the larger communities. One measure of this decline in proportion is provided in the medians for the three groups of colleges at the foot of the column: for the lowest third in population it is 22.2 for each 1,000 of the population; for the middle third, 5.2; and for the highest third 1.9.

In interpreting the figures in the last column it must be kept in mind that they represent the number in each 1,000 of the population of the municipality represented by those students who come from within a radius of 25 miles (including those from the municipality). This column shows the proportion much enlarged for the smaller communities. The decline is much more rapid in proceeding from the smaller to the larger communities than in the column just examined. The medians for the three groups of colleges are, respectively, 26.9, 9.0 and 3.1 per 1,000 of the population.

Other Findings of the Study.-Before proceeding to point out the probable bearing on the junior-college problem of the facts presented, the major findings of some supplementary lines of investigation will be briefly stated without giving in tabular form all the data from which they are drawn. They are as follows:

(a) The percentages of the total enrollment of students in all four years in these colleges who are from the local community, from within 25 miles, from within 100 miles, and from within the state of location are approximately the same as the percentages found for the enrollment in the first two years. The median percentages for the four distances are, respectively, 26.3, 41.7, 70.9 and 86.8. A comparison of these figures with the corresponding figures in the table will show no significant differences. Much the same situation obtains also for the medians, not presented here, of the lowest, middle, and highest thirds of the group of colleges.

The median percentage of students in the last two years who are from the local community for the entire group of colleges is slightly larger than for the first two years and

for all four years, being 29.7. This is evidence of some tendency for the upper classes to be drawn more largely from this source. However, the original distribution of percentages, not given here, shows this larger percentage to be usually restricted to the colleges in the smaller communities. This is indicated by the medians of the lowest, middle and highest thirds, for which the median percentages are, respectively, 25.2, 29.7 and 37.5. While the median for the colleges in the smallest communities is considerably larger than the corresponding percentage in Column 4 of the table, those for the middle and highest thirds are approximately equal to their corresponding figures. The cause for the relatively small influence of these higher percentages in the upper classes upon the percentages for the entire student body is to be found in the fact presented under (c) below.

(b) For all the group of strictly coeducational colleges, i.e., when the data for Numbers 8, 24, 27 and 39 are omitted, the numbers of men and women are nearly equal. This is true for the junior-college years, in which the total enrollment of men is 4,138 and of women, 4,201. These figures represent percentages, respectively of 49.6 and 50.4. It is true again for the enrollment in these years from the local community only, in which case the number of men is 1,143 and of women 1,142. It is not quite as true of the enrollment in the last two years, in which the total numbers of men and women are, respectively, 1,630 and 1,849, and the corresponding percentages, 46.8 and 53.2. This difference may be due solely to the smaller proportion of men who were enrolled in freshman classes during the war period.

(c) An interesting situation is presented by the percentages of the total student body who are found in the first two classes. The median percentage of the total enrollment in these classes is 70.9, the range being from 62.8 to 80.8. This does not represent the normal situation, as, no doubt, this proportion is exaggerated by the unusually large freshman classes for the year 1919-20. But it does emphasize the fact that at the present time the

collegiate enrollment consists very largely of freshmen and sophomores.

PROBABLE MEANING FOR THE ESTABLISHMENT OF JUNIOR COLLEGES

Before it is possible to note the bearing of the facts presented for the problem of proper location of junior colleges it will be necessary to accept some minimum standard number of students that will make feasible the provision of junior-college work. To this end the standard of a total of 50 students in both years set by the North Central Association will be tentatively accepted, although we have little more than empirical assurance that this is an acceptable minimum. The situation may also be tested by a somewhat smaller minimum of 35 to 40 students, which is, roughly, the minimum found in public junior colleges that have survived several years of experience with the plan.

A significant relationship to the problem of feasibility of establishing junior-college work is discoverable in the trend of the proportions of the enrollment in junior-college years from the community of location. The proportion, as has been pointed out, tends to increase as the population increases. It should be apparent, therefore, that the larger the city, the greater the likelihood that the minimum student body will be at hand. Although this is an important consideration in establishing private colleges, it pertains especially to public junior colleges, since enrollments in the latter have been more predominantly from local sources. In fact, they have been almost strictly local as to source of student body. But it will be essential to give careful consideration to the size of city in establishing either type.

We may now turn more directly to the particular size of community in which the establishment of a public junior college is likely to be practicable. The median institution in Column 8 has an enrollment of 51 from local sources, meeting the North Central standard mentioned above. This means that the number from the locality in about half of the institutions falls below this standard of

feasibility. Ten of the 13 institutions in the lowest third when arranged by size of population fall below this point of feasibility, but only 7 of the middle third do so. Describing the situation in another way, we may say that, of the institutions located in the 20 smallest communities, only 5 could meet the North Central standard, if they depended solely upon a local constituency as the public junior college is obliged to do almost entirely. On the other hand, 16 of the 20 largest could qualify in this respect. These figures show that the point of feasibility, roughly speaking, seems to be in cities with populations not often below 8,000 or 9,000. There will be some communities with smaller populations in which it will be possible to secure the requisite number, but there will also be, on the other hand, cities of even larger population in which the establishment of public junior-college work will not be practicable.

Not far from the same outcome is apparent when we apply the North Central standard to the proportions in Column 10. For the entire group the median number in each 1,000 of the population is 5.8. If we apply this proportion to the census figures it appears that it would require a population of 9,000 or thereabouts to make the establishment of a public junior college feasible. It would be uphill work to secure the requisite number of students in most smaller communities. Wide variation in the proportions leads to the same qualifications made in the foregoing paragraph.

The lower minimum of 35 or 40 students could be attained in cities of much smaller populations. There is need, however, of applying this standard conservatively to the present data when considering the establishment of public junior colleges, owing to the greater identification of the interests of the smaller communities with the interests of their private colleges. The smaller the community the greater the identification of interests. This identification of interests could hardly be as characteristic in the early period of the establishment of public junior colleges,

as it has taken many decades to attain it for the private colleges. A conservative conjecture would place the minimum population for the probable attainment of the lower standard at 5,000 to 7,000.

It is not uncommon for parents of young people who desire to attend a higher institution to move into the community in which it is located, establishing residence there. This increases the proportions of the population of the community attending beyond what they would normally be. There is no reason to believe that this will not be characteristic also of communities in which there are public junior colleges. The practise is already found in communities with estimable public high schools.

As intimated above, the private junior college might draw more largely than would the public junior college from the territory beyond the confines of the city of location. This would hardly be true if dormitories were not provided, if denominationalism, the tradition of private higher education, or some other of the forces associated with the private college were not operative. For the same reason, if dormitory facilities come to be provided in connection with public junior colleges the latter will attract larger proportions of students from outside the immediate community than they have drawn to date. It is impossible to predict the exact extent of influence of such a factor upon the proportions drawn, but there is little or no occasion to doubt that it would attain and sometimes exceed the figures for private colleges as presented in the table.

The fact that a relatively small proportion of the students in these private higher institutions are drawn from outside the state seems to many to be a justification for asking the state to foster junior colleges as parts of the public-school system and to encourage attendance upon them from outside the community of location by the provision of dormitory facilities.

UNIVERSITY OF MINNESOTA, MINNEAPOLIS

LEONARD V. Koos