DOCUMENT RESUME ED 403 013 PS 024 268 AUTHOR Glenn .

some of these results were summarized in a previously published article (Glenn, 1989). Cycle 2assessment was conducted in January, 1990 and included the original Montessori sample andadded additional Montessori students. Cycle 3 assessment was done in January, 1993, andincluded all Cycles 1 and 2 Montessori students and added additional Montessori students. Cycle4 was conducted in January, 1996, but no new participants were added.Over the decade of assessment, selection bias has become more and more of a possible problem.A total of 198 participants have been members in the LAS at one time or another. From Cycle 3to this assessment (Cycle 4), no additional members were recruited, and sample fell from 145 to82 (a loss of 43%). Table 1 illustrates that 30 participants have been assessed in all four cycles,23 were assessed for three cycles, and 29 were assessed for two cycles. By status, there were 42of the 82 active LAS participants (51%) still enrolled at FMES, with 49% having moved on toother settings. Selection bias can only be considered possible; we have no evidence that dropoutsdiffer in any significant way from continuing participants. Three equally valid hypotheses exist.First, dropouts may not be any different from participants. Second, dropouts may be resistant toresearch in general and still not differ from continuing participants in 'Montessori' ways. Third,dropouts may be less 'Montessori-like' and discontinued participation due to this lowermotivation. Taking the cautious route, we must assume dropouts did differ from continuingparticipants, making the results herein a 'good' case scenario.Every three years participants, their parents, and their teachers complete a survey. In addition, apersonality measure is administered to the participant, and achievement test results are included ifavailable. Participants no longer at the Earth School receive their packet by mail, since all itemsare self-administered, and all return postage is pre-paid. Privacy envelopes are provided to theparticipants in order to encourage honesty. Participants still at FMES are assessed on a class byclass basis. All students complete their materials on a self-paced basis.Since the beginning of the LAS, the following 9 surveys and versions have been developed: Parent(current and former combined), teacher (current FMES, non-FMES through high school, nonFMES college professor), student (current upper elementary, current junior and senior high,former upper elementary, former junior and senior high, former college and work edition).This is the first year in which some participants were adults, over age 18. For these adults (n 6),the parent survey was of course not administered, but an expanded 'College and Work Edition'participant survey was developed, and a College Professor survey was created as well.The personality measures were selected to be appropriate for normal and healthy persons, andbecause different versions from grade 1 through adult were available. They use common scalesand were developed using a common personality theory (Cattell, 1957).Due to constraints of time and budget, it has not been possible to maintain a comparison orcontrol group. However, in compensation, some results in the LAS can be compared to nationalor published norms. These include the achievement tests and personality measure data. In78

addition, the teacher surveys of participants no longer at FMES ask the teacher to compare theLAS student with other students in the class. There were 20 of the 34 non-adult former FMESparticipants (59%) who submitted teacher surveys.RESULTSWe will find that the primary hypothesis, that number of Montessori education years (MEY) ispositively related to those qualities which are emphasized in the Montessori teaching environment,has not been supported. However, we will find that the secondary hypothesis, that participantswith any Montessori education are as successful as the general population, was stronglysupported and in fact exceeded. As the primary hypothesis has not been supported, presentationof results will be brief. Please refer to the tables for the full numerical results.Dgrn.Qgrapkidngng.hts. Fifty-seven percent (57%) of our 82 Cycle 4 participants were female.Nearly all parent surveys were completed by the mother. Over two-thirds of participants' parentsreported being married, the majority of mothers and fathers had completed college and somegraduate school, the majority of parents wanted their child to finish graduate school, butsomewhat less than half expected their child to finish graduate school. Number of Montessorieducation years (MEY) ranged from 2 to 15, with a mean of 8.39 and a median of 8.0.Participant age ranged from 9.51 to 20.48 (as of January 1, 1996), with a mean of 14.19 and amedian of 14.24 (Table 2 through Table 5). Excluding the 6 adults (mean age 18.57), mean ageof participants still at FMES was 13.00, and the mean age of former FMES participants wassignificantly higher at 14.88.I I. Parents reported that it was somewhat easy to talk to their9. Achildren about the school day or about feelings, the student's ability to cooperate with peers wasrated as good and ability to cooperate with adults was rated as very good, school attitude wasreported as somewhat positive, parents were more likely to report their children as being leadersthan followers, the average number of reminders to do chores was between 2 and 3, and parentsreported a mean of 88 minutes of chores per week for their children (range 0 to 300 minutes,median 60). See Table 5. Parents were asked to describe their child with one adjective.Although no pattern was evident, a listing (by MEY and status) can be found in Appendix A. Inaddition, three parents volunteered other comments (Appendix B).There were no significant differences of parent reported variables in terms of participants status(still at FMES versus gone to other settings).ItarAtunaiskaighlighti. Although teachers reported student subject area of best and secondbest performance, response was quite diverse. Slight standouts were seen in math as bestperformance area and science/computer as second best performance area (Table 6). Teacherswere asked to choose one adjective to best describe the student. Although no pattern wasevident, the list can be found in Appendix C, sorted by MEY and status. Also, 13 teachers of89

former FMES participants volunteered comments, and these can be found in Appendix D.Of particular interest are the results reported in Table 7, teacher reported scaled variables.Teachers were asked to compare the participant with other students of the same gender andsimilar age in their classes. Means and medians from all 15 rating scales were above average, onthe positive side of the scale. This alone may not mean much; the real question is how nonMontessori teachers rate former Montessori students. For all 15 rating scales, means were higherfor teachers of former FMES participants (in comparison with the current FMES teacher ratings),and seven of these were statistically significant (four more were nearly significant). The tablebelow summarizes these significant and near significant differences, sorted by significance level.VARIABLE ( EACHER ASSESSED)AT FMES4.12Abil. to cooperate w/ teachers (1 poor; 5 good)3.68Overall self-image (1 poor; 5 good)Overall academic rating (1 below; 5 above ay.)3.90Appropriate use of spontaneity (1 rarely; 5 often) 3.543.32Ease of distraction (1 easy; 5 hard)4.00Showing grace & courtesy (1 rarely; 5 often)3.90Abil. to finish product (1 poor; 5 good)Showing compassion for others (1 rarely; 5 often) 3.784.02Abil. to work alone (1 poor; 5 good)3.72Abil. to handle stress (1 poor; 5 good)3.90Abil. to cooperate w/ peers (1 poor; 5 good)GONE COMBINED .006.008.022.046.050.061.068.074.080That all 15 scales trended in the same direction and that 7 were significant and 4 nearly so, isstrong support that former FMES students are doing not only as well, but better than, their nonMontessori educated counterparts. This is strong support of the secondary hypothesis.Eanicipantgruidefilighlights. About a quarter of participants reported their best subject area tobe math, and another quarter reported their best area to be language arts/English. Nearly twothirds said they want to finish graduate school. Participants recorded the best job they hope toget, and these were categorized according to Holland's (1973 and 1979) six groups. Given thesample size, roughly equal numbers chose occupations in the Investigative (e.g., scientist, manyprofessionals), Artistic and Social areas (Table 8).Participants were asked to choose one adjective that best describes themselves. Although nopattern was evident, results can be found in Appendix E (sorted by MEY and status). Participantswere also asked to choose a single adjective which best describes their feelings about school, anda pattern Ras evident (Appendix F). Responses were roughly coded as clearly positive (anythingbetter than 'good') or not. While there did not appear to be any difference by age group, roughly9

a quarter of adjectives chosen by participants still at FMES were clearly positive, compared toabout one of 10 for former FMES participants. In addition, among former FMES participants,the clearly positive adjectives clustered among the participants with more MEY. This latterfinding suggests that participants with more MEY are better able to find a way to like their nonMontessori school. However, when asked to rate their feelings about school on a 1 dislike to5 like scale, there was no significant difference by MEY or status (full sample mean 3.99).Participants still at FMES completed 8 rating scales, and former FMES participants completed thesame 8 scales plus four additional scales rating test taking ability (Table 9). Self-rated test takingability (of former FMES participants) was good to very good. The means of the other scaleswere slightly above average to definitely above average. Three of the 8 common scales which allparticipants completed showed statistical differences by status, and two more scales were nearlysignificant. The table below summarizes these results, where participants were asked to comparethemselves with others in their class.VARIABLE (STUDENT ASSESSED)AT FMESGONE COMBINED SIG.Quality of finished work (1 worse; 5 better)General test taking abil. (1 poor; 5 good)Time to finish work (1 more; 5 less)Abil. to pay attn. when noisy (1 worse; 5 better)Teacher expectations (1 too little; 5 too .783.323.233.19.000.001.010.067.085These student reported results also support the secondary hypothesis, and again strongly so.Students who attended FMES and left for other schools reported themselves as better off thantheir non-Montessori educated counterparts on most scales. Note that the teacher expectationsscale, although not quite significant, suggests that former FMES students perceive their current(non-Montessori) teacher as expecting too little of them, as compared to FMES students.We asked students to transcribe the non-paying chores they do at home and to estimate thenumber of minutes per week for each chore. However, differences between participants' methodsof reporting minutes and/or number of times per week made summing to a total number ofminutes difficult. However, with some caution, we estimated a mean of 121 minutes of choresper week (range 0 to 380 or 6.33 hours, median 118, Table 10). This of course was significantlyhigher than the parent estimated number of minutes of chores per week (88). Some readers mightfind of interest the list of chores (by MEY and status), found in Appendix G.Table 11 and Table 12 contain results from the six adults in the LAS. Table 11 contains jobrelated variables, but only 3 participants were working. Table 12 contains self-assessmentvariables. The small sample precludes a valid analysis.10

Saving perhaps the most interesting for last, we asked all participants of junior high age and older,"How has your Montessori experience affected who you are today?". Results have beentranscribed verbatim and sorted by status, MEY, age, and for former FMES participants, numberof years since left FMES (Appendix H). Although there were a few negative comments, theoverwhelming majority of comments were positive to very positive, and some exceptionallyinsightful comments were made. These comments, of current and former FMES students, suggestthat, from the student perspective, their Montessori experience was very positive and continues toaffect who they are today. These comments also add strength to the support of the secondaryhypothesis.All LAS participants took apersonality measure. As described in an earlier section, the Cattell series was used, as they areappropriate for healthy people and offer versions from elementary school through adult. All meanscores were within one standard deviation of the norms, supporting (objectively) that LASparticipants as a group do not significantly differ in personality. Also, there were no significantinteractions with MEY, suggesting little effect of a Montessori education on one's personality.There were 70% of former FMES participants who provided copies of grades. For all lettergrades combined, two-thirds (67%) were As, 22% were Bs, 6% were Cs, 2% were Ds, and 3%(n 5) were Fs. The grade point average was 3.48.All participants still at FMES but under high school age took the short form of the MetropolitanAchievement Test. For the total battery, this sub-group scored at the 74th (national) percentilecompared to national norms.Some participants no longer at FMES and under high school age (11 of 34 or 32%) submittedcopies of achievement test results. Of course a variety of tests were used, but this sub-groupscored at the 94th percentile compared to usually national (sometimes regional) norms.High school aged and older participants (at FMES or no longer there) submitted PSAT or SATscores. Mean verbal for current FMES participants was 81st percentile, and mean math was 73rdpercentile. For former FMES participants, the mean verbal was 78th percentile, and the meanmath was 69th percentile.Since it may have been likely that higher scoring participants were more likely to submit gradesand achievement test results than other participants, these results should be considered a best casescenario.CONCLUSIONThe primary hypothesis that the number of Montessori Education Years is positively related tothose qualities which are emphasized in the Montessori teaching environment cannot be1112

supported. However, the secondary hypothesis that participants with any Montessori educationare as successful as the general population was strongly supported. Actually, saying 'any'Montessori education is really not accurate, because 98% (all but 2 participants) had at least 4MEY. Objective measures (ratings by current non-Montessori teachers of former Montessoristudents, personality inventories, achievement test results, and grades) all support the conclusionthat any (4 or more years) Montessori experience is beneficial and caries over to other educationalenvironments.THE FUTUREBecause of consistently finding little support for the primary hypothesis (but strong support forthe secondary hypothesis), the LAS will be moving to a more qualitative approach. Also, wefound this cycle that several participants, particularly the older ones no longer at FMES, wereresistant to investing the one to one-and-a-half hours necessary to complete LAS materials.As such, the LAS will probably move to a 15-20 minute interview format (in person if still atFMES, by phone if gone). Beyond the general 'how are you doing?,' we will be asking guides,parents, students and administration officials associated with FMES for suggestions as to what tocover in the interview. We hope to regain many of the LAS dropouts with this streamlinedformat. It is also possible that junior high aged FMES students may do the actual interviewing aspart of their Business Discovery curriculum. Beyond learning interviewing techniques, persons inthe middle of adolescence may find it helpful to hear others' perspectives, particularly those offormer students who are now adults.12

REFERENCESBoehnlein, M.M. (1985). The NAMTA Montessori bibliography - A bibliography of sources inthe English language; 1909-1984. NAILLA Qmagrlydp (2).Boehnlein, M.M. (1986).k ALUI0 it.L. g.110 .1.:Cleveland heights, OH: N.A.M.T.A.Boehnlein, M.M. (1988). Montessori research, analysis in retrospect. TheJg81611Aiourna,11(3).Boehnlein, M.M. (1994). The NAMTA Montessori bibliography A bibliography of sources inthe English language; 1909-1993. Second Edition. hi,A,M1Aacturnal,i9 (3).Cattell, R.B. (1957). assonalityandingliyatiglutrwurcandmosuremmt. New York:Harcourt, Brace & World.Glenn, C.M. (1989). A comparison of lower and upper elementary Montessori students with apublic school sample, TheNAILTAIonmal, .14, 63-68.Glenn, C.M. (1993). Market research at a Montessori school: Reasons for choosing, staying, andleaving, Monttssoragg, 5, 13-14.Glenn, C.M. (1994).Portland, OR: Franciscan Montessori Earth School. (ERIC Document Reproduction Service No.ED 370 679).Englewood Cliffs, N.J.:Holland, J.L. (1973).Prentice-Hall.Holland, J.L. (1979). The jsthsfinda Palo Alto, CA: Consulting Psychologists Press.134

TABLE 1PARTICIPANT STATUS BY YEAR OF FIRST ASSESSMENTTOTALLAS PARTICIPATION STATUSStill At FMESCOUNTPERCENTNot At 2100.0%Started in 1987Started in 1990Started in TUDINAL ASSESSMENT STUDY: CYCLE 4 (TEN YEAR FOLLOW-UP)1996 Franciscan Montessori Earth School - Christopher M. Glenn, Ph.D.

TABLE 2PARENT REPORTED CATEGORICAL %3.7%15.9%2.4%9.8%TOTAL82100.0%PARENT REPORTED HIGHEST LEVEL OFEDUCATION WANTED FOR STUDENTSome CollegeFinish CollegeSome Graduate SchoolFinish Graduate SchoolNo 102.4%4.9%30.5%7.3%42.7%12.2%82100.0%WHO FILLED OUT SURVEY?MotherFatherMother and FatherNo responseTOTALMARITAL STATUSMarriedSeparatedDivorcedUnmarried, Living WithNo response.PARENT REPORTED HIGHEST LEVEL OFEDUCATION EXPECTED FOR STUDENTHigh SchoolSome CollegeFinish CollegeSome Graduate SchoolFinish Graduate SchoolNo responseTOTAL6131324474256LONGITUDINAL ASSESSMENT STUDY: CYCLE 4 (TEN YEAR FOLLOW-UP)1996 Franciscan Montessori Earth School - Christopher M. Glenn, Ph.D.

TABLE 3NUMBER OF MONTESSORI EDUCATION .00Sample82Minimum Maximum2.0015.00LONGITUDINAL ASSESSMENT STUDY: CYCLE 4 (TEN YEAR FOLLOW-UP)1996 Franciscan Montessori Earth School - Christopher M. Glenn, Ph.D.

TABLE 4AGE OF d)LONGITUDINAL ASSESSMENT STUDY: CYCLE 4 (TEN YEAR FOLLOW-UP)1996 Franciscan Montessori Earth School - Christopher M. Glenn, Ph.D.17

TABLE 4AGE OF 7.1717.3117.3617.451

DOCUMENT RESUME ED 403 013 PS 024 268 AUTHOR Glenn, Christopher M. TITLE The Longitudinal Assessment Study (LAS): Cycle 4 (Ten. Year)