Click for next page ( 8


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 7
WHAT HAS BEEN HAPPENING TO TIME TO THE DOCTORATE? While factors leading to attainment of the doctoral degree have attracted research attention over the last 30 years, only recently has interest focused on the length of time it takes to earn the degree. Surprisingly, most current studies seem to overlook the phenomenon of increasing time to the doctorate occurring over the last two decades. Aggregate data on doctoral degrees show that while median time to the doctoram decreased in the limos, the decline was followed by a rather swift and steep increase through the 1970s and l980s (Figure 1~. Although lengthening degree time might simply reflect a distributional shift from doctorates in fields in which time to the doctorate is short (such as physical sciences and engineering) to those in which it is longer (such as humanities and education), other studies have found the increase is occurring in all fields (Coyle, 1987~. 11 ~ 10 o. e:7 6 5 1958 1962 1966 1970 Total Time : J - - 1974 1978 1982 1986 Year Figure 1 Median years to the doctorate, all fields combined including humanities and education fields, 1958-1986. 7

OCR for page 7
Components of Time to the Doctorate and How They Have Changed Through Time The Several Kinds of Time The time required to complete the doctorate can be measured in a number of ways, and the Me of measurement used affects the degree of observed change. The most comprehensive measure of time is total time to the doctorate (loll)), defined as the time from receipt of an undergraduate degree to completion of the doctorate. I-11) is particularly useful for "pipeline" studies that examine the availability of new doctorates to enter the labor force. Similarly, 'I ~ is useful for determining how quickly the supply of doctorate-level personnel will respond to changes in the demand for people with doctorates. Other things being equal, for example, a 10-year TTD would mean a delayed response of new doctorates to an increase in demand and a long wait for employers wanting to hire them. change as well as conclusions about which factors led to that ~ . Time to the doctorate also can be measured by the length of time that a student is actually registered in graduate school. Registered time to doctorate (RTD) is defined as TTD less the length of time prior to graduate entrance (TPGE) and any other lime not enrolled in the university (THEIR that is, RTD = TID - (TPGE + TNEU). TPGE may consist of service in the armed forces, time spent in travel, leisure or home-related activity, and/or postbaccalaureate work experience. There are two additional elements of RTD for which we have no measure: time spent in actual study/work toward the degree and time spent at the university in other pursuits. RTD is not a measure of the minimum time needed to complete the doctorate, since time spent in nondoctorate-related activity is also included. RTD, like LID, is a measure of how quickly supply can respond to demand. In addition, it can be used as an indicator of the need for faculty and other resources in a graduate program. The relationship among these four time measures is summarized in Table 1.1. Mean TTD for each of 11 science and engineering fields-chemistry; physics and astronomy ("PEAS; earth, atmospheric, and marine sciences (SEAMY; mathematical sciences, including computer and information sciences ("math"~; engineering; agricultural sciences; biological sciences ("biosciences"~; health sciences; psychology; economics; and all other social sciences ("social sciences") is taken from the Doctorate Records File (DRE9, the data base of the Survey of Earned Doctorates conducted annually by the National Academy of Sciences' Office of Scientific and Engineering Personnel (see, for example, Coyle, 1987: Table 2~. Mean TTD, rather than median TTD, is used because it is more sensitive to small yearly changes in the data and easier to compare 8

OCR for page 7
TABLE 1.1: The Relationship Between the Several Time Measures Year of Undergraduate Degree Completion + Time Spent Prior to Graduate Entry (TPGE) = = Year of Enhance into Graduate School Time Spent at die University Working on Degree or Other Pursuits (RTD) Time Spent Not Enrolled at He University (TNEU) Year of Graduation with a Doctoral Degree (MOD) among fields.1 Although mean values can sometimes be distorted by the existence of a few outliers in the data, we did not encounter evidence of this problem (see Appendix Tables 2.1-2.5~. The time required to complete the doctorate has been increasing in the sciences and engineering primarily because students are spending more time in graduate school (i.e., RID is rising). Figure 2 contrasts the growth of RTD with changes in its component measures, TPGE and TNEU.2 The effects of changes in the intervening years are explored in the next section. Mean Total Time to the Doctorate Mean ,1-1 D increased in each of the 11 fields from a low of about four months in economics to a high of nearly three years in the health sciences (see Appendix Table 2.1~. All but biosciences and agricultural sciences experienced double-digit percentage increases in 'l-1~. The greatest increase, 30 percent, was in math, and 1-1 D lengthened significantly even in fields in which it already was quite long. For each field, the within-year variation in 1-1 D decreased from 1967 to 1986, suggesting student completion times more concentrated around the mean. 1 Means are also used to provide the estimates of person-year losses shown on pp. 22 of this chapter. Although our analysis is confined to a discussion of mean times, median times have also been increasing (see Appendix Table 1~. 2 Appendix Tables 2.1-2.5 display the mean Ads, TPGEs, RTDs, and TNEUs and their respective standard deviations, for each of the 11 fields at two points in time: 1967 and 1986. 9

OCR for page 7
O i6 - ~Chemistry P&A 14 12 u, 10 8 6 ~ 4~ cat 2 o ^_ ~ . ~ ,, ~ /,, ~ f ~ ~ 1 ~ C ~ O- _ T 1967 1986 EAM it' 1967 1986 . 1967 1986 Year - :~ 18 o o - 4 _ 12 o 10 Math/Comp Engineering Agri. Sci. 8 6 4 2 0 1967 1986 ~ , . i 0 it. r _ , . ~ T ~_ . . ~ 1967 1986 Year 0 TPGE 0 RID ~ . 4 1967 1986 Figure 2 Components of mean total years to the doctorate, by field, 1967 and 1986. 10

OCR for page 7
~ 2 - o - o 18 16 l4 12 is 10 8 6 4 _ ~ es o ~ ' 18 ~ O 16 14 ~ 12 o ~ 10 s" ~ O r: I 1967 1986 . Biosciences Hewn Sci. Psychology f , /. . 1 ~ . 1967 1986 Year Economics CKh" Social Sci. ~ ~ o 1967 1986 , 1967 1986 a, , I' 1967 1986 . 1967 1986 Year 11 TO 'l TPGE - t~ RTD

OCR for page 7
Mean Time Spent Prior to Graduate Entrance Changes in mean time spent prior to graduate school appear to have had little impact on the rise in TTD. TPGE showed little change. Except in health sciences, where there was an increase of approximately one year, on average, students in chemistry, P&A, EAM, and math entered graduate school less than one year after completing an undergraduate degree. Those in engineering, biosciences, agricultural sciences, psychology, economics, and social sciences spent between one and one-and-a-half years before entering graduate school. In health sciences, mean TPGE was a little over two years. Although fairly large TPGE increases occurred in three fields math, psychology, and health sciences TPGE was a small portion of TTD in most fields. Two fields EAM and agricultural sciences-experienced a decrease in TPGE. Analysis of the coefficients of variation for each year again revealed that within-year variance went down between 1967 and 1986, suggesting greater concentration of TPGE times around the mean. Mean Registered Time to the Doctorate Ideally, registered time to the doctorate should be broken down into time spent working toward the doctorate and time spent at the university in teaching or other activities unrelated to the doctorate (Berelson, 1960~. Unfortunately, the DRF does not separately identify these two components. In all of the fields in the study, RTD increased at double-digit rates (see Appendix Table 2.3~. Measured in both percentage and absolute terms, the largest increases occurred in the social sciences (where RTD rose from 5.9 to 8.8 years, or almost 50 percent) and economics (where RTD jumped from 5.1 to 7.0 years, or 37 percent). The smallest increases in RTD were in chemistry, P&A, and engineering. Overall, increases in RTD accounted for at least half of the increase in TTD and, in some fields, it accounted for over 100 percent of the increase.3 Meat' Time Spent A way from the University Students have many reasons for leaving the university prior to completing the doctorate. They may have financial difficulties, may be discouraged and/or frustrated with academe, or may need to seek additional data to finish the doctoral thesis (Dolph, 1983; Spady, 1970~. Time not enrolled in the university increases TTD and, hence, is a variable worthy of separate 3 This happened because decreases in the other components brought 1TD down. 12

OCR for page 7
consideration. In most fields, TNEU decreased by at least half a year between 1967 and 1986. However, there was wide variability among the fields. For example, the decline was 1.5 years in economics, almost a year in health sciences, half a year in the biosciences, two-and-a-half months in math, and less than a month in psychology. Within-field variation for lNEU decreased in six fields and increased in five (see Appendix Table 2.4~. Summary The major factor responsible for the change in '~-1D between 1967 and 1986 was the growth in RTD. In a majority of fields, a decline took place in TPGE and in INEU (see Appendix Table 2.5~. The Nature and Significance of the Time Trend The literature suggests it is now taking longer to complete a doctorate than at any other time. The upward slope of ITD follows a rather extended period of stability in time to the doctorate. In the near future, it will take even longer for doctoral candidates to complete their degrees. The Two Models The authors used statistical modeling to look at changes in 1TD and other variables during each of the years between 1967 and 1986. For each field of study, regression equations were estimated using TTD or one of its three components as the dependent variable and time as the independent variable. Two different models, one which assumes that time has a linear effect [AID = f(T)] and another which assumes a non-linear effect over time [TID = f(T,T2~l, were used. Using the linear model, for example, for chemistry students resulted in the conclusion that TID increased by an average of 0.03 years per annum (or roughly 1 1/2 weeks per year) during the 1967-1986 time period (Table 1.2~: a chemistry Ph.D. in 1967 took one-and-a-half weeks longer to complete the degree than in 1966 and nearly 30 weeks longer in 1985. Using the non-linear model, the increase in TTD for a chemistry doctoral candidate was about three weeks in 1966 and about 62 weeks in 1985.4 4 These figures were determined as follows: the increase from 1966 to 1967 = 0.0632 years = (0.065 x 1) - (.0018 x 5) and from 1966 to 1985 = 1.20 years = (0.065 x 19) - (.0018 x 19~. 13

OCR for page 7
The non-linear model produces a larger annual increase in ICED over time than the linear model for most fields and, with the exception of agricultural sciences, values derived from both the linear and non-linear models are statistically significant. In general, the non-linear model explained more of the variance than the linear model and, in most fields, provided a better fit of the data, hence a more accurate estimate of the effects of time on LID. Patterns of Charge i] In all 11 fields, there is a distinct and statistically significant upward trend n both TTD and RTD (Tables 1.2 and 1.3), although the trend is more pronounced for RTD than for TTD. For TTD, a non-linear time trend exists in most fields, suggesting that both the increase in time to the doctorate and the rate of change have differed across fields. Completion times accelerated in seven fields (EAM, agricultural sciences, biosciences, health sciences, psychology, economics, and social sciences) and accelerated and then decelerated in four (chemistry, P&A, math, and engineering). For RTD, distinct patterns also emerge for each field, with some showing acceleration and others showing deceleration. A comparison of RTD and TTD suggests that in most fields the coefficients are quite close. This is not the case for the other components of time to the doctorate, however, suggesting that RTD is the factor most responsible for lengthening HID. An examination of time trend coefficients for the set of regressions using TPGE as the dependent variable shows that, in all fields, the amount of variation explained by time is less for TPGE than for RTD, in some cases half as much (Table 1.4~. The non-linear model is preferable to the linear one in most fields, although in some fields its use has little impact on R2. Using the non-linear model dramatically improves fit in the biosciences, economics, and social sciences; and it shows small gains in R2 in math, engineering, health sciences, and psychology. The results again suggest that the time trend differs among fields. The final set of regressions uses TNEU as the dependent variable (Table 1.5~. In the linear model, mean time not enrolled in the university falls in seven fields (chemistry, P&A, EAM, agricultural sciences, biosciences, economics, and social sciences); rises in math, health sciences, and psychology; and remains stable in engineering. The results again suggest that the non-linear model provides better predictions of the time component in most of the fields. 14

OCR for page 7
e, o c) . - - - - ~ - E~ ~ - u, E~ u, o .- c~ ~: o v c) * c~ cs ~ e ~ ~o ~ v * ~: - o ~ a: o o o ~ oo c~ o oo o o o o o oo O en O O O O O ~ ~= ~= ~c~ ~= ~~ o ~ o ~ o o o ~ o ~ o ~ o ~ o ch o ~ o c~ o ~ o ~ o ~ o ~ o . . . . . . . . . . . o ~ o ~ o ~ o ~ o So o ~ o o o I - I _1 - - I _ _ ~t - - - ~= ~ ~ ~o o ~ ~ ~ oo ~ o ~ ~ o o c~t ~ ~ o ~ o o ~ oo o . . . . . . . ~ o~oc~ol ~o~omo - - N ~oo ~ Go o ~o o . . . ~oX Soo ~oo ~o ~o o o o o o ~oc ~o~ ooo ~c~ .....~ oooooo o ~o ~ ~oo~o c~ ~ ~o ~ c~o~o o . . . . ~o~o o oo ~ - ~ - ~ c~ ~ oo ~ ~ ~ ~ N . . . . - c~ o ~ o ~ o ~ o - l ~ l ~ l ~ l - - - . oo v ~ o ~ ~oo ~D Co . . ~ ~ o - ~oo ~oo O ~D o o o o o o ~ - - o ~ - ~ - . = 0 e~ 0 ~ ~ o o o oo 4- - ~- ~ o ca c~ .O = ~ ~ -~' ~- ~ ~ ~ cr~ ~ ~ ~ ~ V~ ~ 0 ~ ~ ~ ~ e~ ~ ~ ~ ~ c~ o ~ o ~ o ~ ~ ~ o ~ o ~ o ~ ~ o ~ o . . . . . . . . . . . . . . . . . . . . . . o ~ o ~ o ~ o c~ o ~ o o - o ~ o ~ o ~ o ~ o ~ c ~oo~ oo ~o~ ~oo ~Y ...... ... .. oo ~oooo ~r ~oo ~c) ,, ~ 5? ~ .~ ~, ~ 80 E: m tc 15 o ._ - ~ ~c =: _ ~ .s - ca ~D o .= cp cd ~D ct, x ._ _ - - _ u, .= ~ ~ ,D c~ .= * ~

OCR for page 7
- o ~ - o - ~ .g E~ u, ~- oo p~ ;~ cq ~> oD e~ ~- ~ - cn e E~ c~ E~ -* - ~o ~ 'e o~ ~o * c~ ~ E~ o ou ~ cd o o ~ - ~D ~O a ~o~ ~a ~a~ ... ...- oooooo ooooo e~ _= ~ .oo ~ _ ~ _ ~ _ ~ ~ ~ _ O ~ ~ _ ~D _ \0 _ ~ _ ~ _ O ~D O 00 0 ~ O ~ O ~ O ~ O ~ O \0 0 ~ O 00 0 o ~ o ~ o ~ o ~ o ~ o o o ~ o oS . . . . . . . . . . . . . . . . o ~ o ~ - o ~ o ~ o ~ o o o ~ o ~ o oS o ~ o o 1 _ 1 ~_ 1 _ 1 ~_ ~001 _ _! - - - - - - ~- d-~~ So o o, Y o ~ ~ ~o o ~ ~ - ~ ~ ~ o ~ o o o ~ o ~ ~ ~ o ~ o ~ o ~ o v~ o ~ o ~ o ~ o o o ~ o ~ o ~ o oo - da ~_ 1 _ 1 _ I _ ~_ - - - - - S ~o ~In 0 ~ o ~c ~oN ~oo .. . In S cr ~r~oo ~oo ~o~ ... .. . oooo ooooo o o So - - ~ - - ~ - ~ - ~ - o - oN ~D ~ ~ 00 ~ ~ ~ ~ ~ ~ ~ \8 0 ~ ~ C~ ~ ~ In o ~ o ~ o ~ o Yo o Yo o ~ o ~ o c~ o ~ o ~ ~ oo ....... o~o~omo~o~ooomo~ooo~oo - - ~- -l - - ~ - - - - - - - ~D ~\0 0\ C ~C ~00 ~ l ~1 sD O \0 ~- ~) ~t- d ~ ~ ~u ~In ~ ,= ~ E ~ ~ ~ .', ~ o E ~ o -= 16 ._ o ;> o _. ca C~ .= .~ o - e~ ~s .f U, D ~ ~D O ~ ~ _ :' ~ U, ._ "C td U) _ .= ~ ~.D C~ .= * ;>

OCR for page 7
v o e = ED .6 a EN .g Ed - * - ~o ~ to I ~ o o lo ~ ~ co - o c, a o .O Cal ~ ~000 ~00 ......... OOOOOOOOOOO c) O ret r" ~to ~ O ~0 00 0 _~ ~ ~0 ~ - ~ - ICY ~ ~ - ~ O ~ O O O ~ O ~ O ~ O ~ O ~ O ~ O TO ~ O O~OOO=O - O.=O=O=O=O - O - OF O O O O O O O ') O ~ O O O ~ O ~t O ~-O ~ 0 00 _ _ _ _ ~ _ _ ~ ~. ~ ~ _ - - - O ._ \0 C~ ~.= ~ ~<~ ~ O'` ~ ~ ~ ~ ~ O _ O ~ O ~ ~ ~ O t_ ~ ~ ~ ~ ~ ~ ~ oo C~ O O O O t-O ~-O O O w) O ~ O ~ O ~ O ~ O C~ O ~ .... 4- O O O O O O O O O ~ O ~ O ~ 0 ~4 0 - '0 \0 0 C ~C7 _ _ 1 _ 1 _ _ 1_ 1 ~1_ 1_ 1 _.1 C~ _ _ _ C ~C ~C~ ~D ~4 ~00 oo ~ . ... ... . O O OO O ~O ~O C ~00` 0 C ~ O d O~- O ~t ~oo O . . .. . . . . O O OO O OO ,0 0 0 0 C~ ~t- ~ ~ ~m ~O C~ ~ ~O ~ - ~C~ ~ ~ - ~C~ ~ _ O ~ O ~ O oo ~ ~ O ~ ~ ~ ~ ~ ~ ~ ~ ~ O ~ ~ ~- O C~ O ~ O O O ~ O ~ O ~ O ~) O ~ O ~-O ~ O C~ . . . . . . . . . O C~ O ~ O O O oo O O O ~ 'O ~ O ~ O oo O O O C~ _ _ _ _ _ 1 _ _ __ 1 _ _ ~O ~ O ~o oo ~o ~0 c ~O ~e~ ~oo ~0 ~r~ ~0 . .. . ~. o o o o ~o ~o o ~ .t om ~ o ~ E ~ E ~ ~ -= ~ ~ ~ c V 17 U, _d .= ._ o ._ - Ct .= _ .= ~ . ~ ~V ._ _ ;> o ~ ': V ~ P.. ._ ~ C~ _ .= ~ ~ .= .= * ~ 5 . - x C~ ._

OCR for page 7
.- c~ o o z > ~ ~c E~ ~ - E~ - n ~: - c~ - v) - ~ E~ c~ O ~o E~ ~ - - ~ o o * c~ - ~ - ~d o - c) o o ~ - - S ~ ~o~ooo ~oooo . .........~ o oooooooooo o ~ - ~Yo ~ o ~ - ~ - ~o ~o ~m ~c~ ~m ~\o o o o ~ o o o oo o ~ o ~ o - ~ oo o ~ o o o o ~ o o o ~ o - o ox o o o ~ o oo o . . . . . . . . . . . . . . . . . . o o o ~ o ~ o o o o o u~ o oo o o o ~ o ) - - - oo ~ = ~o ~-~ ~ ~-~ ~ ~ -~n oo os 0 0 0 ~ ~ oo ~ ~ O ~ ~ ~ oo o ~ o oo o ~ o o o - ~ ~ o o ~ o o c~ ~ . . . . . . . . . . . . . . . . . o o o - o ~ o o o o o c~ o ~ c ~ o oo o _ _ ~ - ~ - 1 - ~ _4 ~ ~ ~ C~ ~ C~ 1 - - - - ~ - So - ~ o ~ ~ c~ ~ . . ~ o o 1 C)O ~- ~000 ~oo ~In0oo ooo ~oo oo ~-Soc ~oo~ ........- oo ~c ~c~e~ ~oo ~ ~c~o~ ..~ oooo o oc~t o ~so . ooo r~oc~ ~ -~o~ ... ooo = ~C5` ~ ~0 ~ `0 ~= ~ ~' ~= ~0 _% ~ C~ O ~ ~ O ~ ~ O ~D ~ e~ ~ ~ ~ ~ ~ ~ In ~ 0 O ~ O iD O ~ O ~-O O O ~ O ~ O ~ O ~ O . . . . . . . . . . . . . . . . . . . o ~ o ~ o u~ o c~ o o o ~ o ~ o ~ o ~ - o ~ o ~ ~_ ~ _ ~_ _ ~ _ ~_ ~_ _ _ ~_ 1 ~ o ~o ~ ~oo - - o o c ~ o o c~ ca u, . . ~;- c) ~ i~ ~ .O , , ~ s o ~ ~ ~o ~ 5? w ~ c ~ N .. oo .e c . - o4 18 ._ C) ~q C~ W 10 2 ~ ~._ - U, ~. _ ~ m 00 ~DC~ 00 ... _C~ CeO q O~ O _C)~ W;-,O 5: ~ ~ '_= ~ ~ 0 O . . ~ O O ~._ _ _ O w _ ;> o o U, C, - c) .= .5 o ,_ Ct - ,_ U, ~ C, ~D ~ w ,~ ;> cn C) ._ C) V, w ._ o V) o .5 ,, W . C~ ._ 4= d_ _d ~3 ~ ~; C~ _ .= ~ W C~ * ;>

OCR for page 7
The Shape of Change The analysis shows time to the doctorate lengthened in all fields, and the time Send was non-linear for each of the three time components that make up HID. Regression analysis revealed two distinct patterns in TTD (Tables 1.2 and 1.3~. Data in eight fields was U-shaped, with a negative (or positive) T and a positive T2 term, leading to an initial rise in TTD or a decline followed by an acceleration in TTD. This pattern existed in EAM, math, agricultural sciences, biosciences, health sciences, psychology, economics, and the social sciences. That AD in these fields may continue to lengthen at an increasing rate over time is a source of potential concern. Data for three fields showed an inverted U shape, with a positive T and an negative T2 term. For chemistry, P&A, and engineering, this pattern led to an eventual decline in the rate of increase in TTD over time. Figure 3 shows the actual data for each of the 11 scientific and . ~ engmeermg tie .c .s. Since the non-linear time-trend model explained more of the variation in I-lD than the linear model, it was used to forecast l-lD for 1987. The results were then compared with the actual AD values for 1987 (Table 1.6~. The non- linear model closely projected ITD in ~ of 11 fields (within 0.01-0.34 year) but underestimated by close to half a year ICED in math/computer sciences, EAM, and agricultural sciences. The model produced a slight overestimate in TTD in the health and social sciences. For engineering, the projected and actual values were virtually the same. Manpower Loss from Lengthening Total Time to the Doctorate One important implication of the lengthening of TID is that a given doctorate yields fewer potential person-years of labor force effort to society. The potential manpower loss calculated from increasing I-lu does not equate to the total social implications of this trend. For example, no allowances are made for changes in the quality of new doctorates, market salaries, unemployment rates, on-thejob training times, or losses of Ph.D. positions at institutions that use predoctorates for research, teaching, or other work activities. Similarly, graduate students who might have been discouraged from obtaining a doctorate because of the time required to earn the degree are left out of the calculation. In addition, the baseline year used for the calculation is 1967. No presumptions are made as to whether the TTD in 1967 was better or worse than that which prevailed in some other year, since the goal is not to define the optimum year on which to 19

OCR for page 7
~ 18 A 16 Go 14 ~ 12 ~ . 10 8- l 6 4- 2 O dy 1967 1986 18 16 ~ 14 12 ~ . 10 ~ 8 6 4 _ 2 _ ~ it' ~ / O- ~ ~ , ~ 1967 1986 Biosciences Heat Sci. Psychology . . _ F 1967 1986 Year Economics Over Social Sci. L T . _ . loo ma ~ ~ . roe ~ ~ ~ . ~ . it' 1967 1986 Year _ - 1967 1986 i! ~ I ~ . 1967 1986 Figure 3 Mean total time to the doctorate, by field, 1967-1986. 20 59 TPGE E RID

OCR for page 7
~ 18i 16 'o ~ 14 ~ ' ~ 12 o _ ~ 10 c~ ' ;,, 8 O 6 : Chemistry P&A ma , ~ 1967 1986 . 1967 1986 Year 18/ 16 -3 14 ~ 12 10 8 6 4 2 0 i, _ 1 . 1967 1986 . ~ ~ : PAM _: ma. it, ~- /~ 1967 1986 ~ - Math/Comp Engineering Agri. Sci. 1 L ~ . ~ ~ , ~: . _ T 1967 1986 Year 21 am me. ~ ~e ~ 7 1 967 1986 gel TPGE ~] RTD

OCR for page 7
TABLE 1.6: Difference Between [forecast and Actual l1Ds, 1987 Field of Doctorate Forecast* Actual Difference Chemistry 6.90 7.05-0.15 Physics/Astronomy 7.94 8.09-0.15 Eanh/AtmosphericlMarine Sciences 9.83 10.26-0.43 Mathematics/Computer Sciences 9.41 9.86-0.45 Engineering 9.04 9.05-0.01 Agricultural Sciences 9.17 9.71-0.54 Biosciences 9.11 9.020.09 Health Sciences 13.81 13.520.29 Psychology 1 1.5 8 1 1.240.34 Economics 10.07 9.740.33 Social Sciences 13.44 13.170.27 *Based on non-linear trend equation. TABLE 1.7: Maximum Potential Person-Years Loss Resulting from Lengthening Total Time to the Doctorate, 1968-1986 Estimated Number Loss as Percent Field of Doctorateof Lost Person-Years of Total* Chemistry1 1, 815 4 1 Physics/Astronomy11,801 6 1 Earth/AtmospherictMarine Sciences3,872 40 Mathematics/Computer Sciences13,306 85 Engineering16,415 42 Agricultural Sciences500 4 Biosciences-17,082 -28 Health Sciences5,529 63 Psychology29,936 62 Economics-1, 8 8 5 -16 Social Sciences8.751 27 Total 82,958 *Determined by dividing "estimated number of lost person-years" by the total number of new doctorates provided during this period. 22

OCR for page 7
calculate T1D but, rawer, to provide a quantitative estimate of how much high- level manpower has been lost over time. To figure manpower loss, mean Lyle calculated for each field and each year is subtracted from the 1967 mean TTD. The result is multiplied by the number of new doctorates in the given year to determine the manpower lost. The total loss for each field is calculated by summing the loss in each year beginning in 1968 and ending in 1986. A percentage loss is calculated by dividing the total person-years lost in all fields by the total number of new doctorates produced during this period. The calculation assumes all new doctorates are employed. Table 1.7 and Appendix Table 3 provide crude estimates of the potential gain in Ph.D. supply if AD was reduced to the 1967 level. It should be noted, however, that these may be upper-limit estimates of the loss because many individuals pursuing the doctorate over an extended time simultaneously performed other work whose value to society cannot be determined. These figures do not take into account the effects of increases in TTD in discouraging career choice. Table 1.7 suggests that a small but meaningful increase in supply greatest in psychology-could be achieved if the trend toward a longer l l L, could be reversed. 23

OCR for page 7