As outlined in Chapter 2 the estimation of the input-output model and the making of projections of the number of students at various stages of the higher education system requires data on:
It is possible to estimate the model with data from just two consecutive years. In this report we use course enrolment in higher education and school enrolment data from 1993 and 1994, and course completions data in higher education from 1993. The population estimates are for 1993 and 1994 and projections for 1995 to 2001. The three data sets are described and their initial analysis reported in the following sections.
3.2 Course Enrolment and Completions Data
Data on course enrolment and completions is used to estimate the matrix of transition proportions. It was obtained from unpublished records kept by DEET. Course enrolment and completions data are stored on separate files as aggregated records. These aggregated files are derived from unit record files provided by each university funded by DEET. DEET releases only aggregated data to safeguard confidentiality of individual students.
Specially compiled course enrolment files for 1993 and 1994 and a course completions file for 1993 were obtained from DEET. These files are different from the files that DEET normally makes available in one significant respect. In the specially compiled files the course commencement date (month and year) of students is included instead of just the year since commencement (one, two, three or three plus).
A vast amount of information is stored on each file. However, not all the variables that are included in the enrolment files are included in the completions file. For example, variables indicating whether a student commenced the course as a school-leaver or not and that indicating the mode of attendance, that is full-time, part-time or external, are not included in the completions file. This means that the analysis that can be performed will be limited by what is contained in the completions file.
The course enrolment files contain information on thirty two variables, and the course completions file on only thirteen of these. At this stage the analysis is restricted to looking at the system at the national level, although the model is equally applicable at the state/territory level. Only the following seven variables are used to estimate the input-output model:
The 1993 enrolment file has over 355,000 records while the 1994 file has over 370,000 records. The 1993 completions file has just under 86,000 records. SAS software was used to read and interrogate the data files on an Alpha machine. A variable giving the year of enrolment in the system since course commencement, Y, was generated from the course commencement date. In order to be consistent with DEET's definition of a commencing student the following definition was used for this variable:
,
where Ref_year is the year of the data file, Year is the year in which the course was commenced and month is the month in which the course was commenced. It should be noted that DEET defines a commencing student as one who is in his/her first year of a particular course at a particular institution. This means that a student who for some reason changes course or begins another degree after having completed one, is classified as a commencing student for a second time even though he/she may not be new to the university or faculty.
Preliminary analysis of the data revealed that students' age ranged from zero to ninety-eight! There is no reasonable explanation than coding error for the existence of students of age zero or 98. There was also some doubt about students who were in their nineties, and therefore, anybody whose age was zero or over 89 years was excluded from the analysis. This process resulted in the removal of 240 students in the 1993 enrolment file, 54 in the 1994 enrolment file and 25 in the 1993 completions file.
A value of negative one for Y in the 1993 course completions file implies the commencement date of between April 1994 and December 1994 for these students. A coding error is the most likely explanation for such a value of Y. There were a number of students whose commencement and completion times were between April 1993 and December 1993 (Y equal to zero). Similarly Y equal to zero in the enrolment file for 1994 indicates that some students' commencement date is after March 1994.
In order to be consistent with the data in the enrolment file for 1993, students for whom the variable Y is less than one are excluded from further analysis. This resulted in the removal of 122 students (109 undergraduates and 13 postgraduates) from the 1994 enrolment file and 598 (300 undergraduates and 298 postgraduates) from the 1993 completions file. Since the number of such students is relatively small, their exclusion is unlikely to significantly affect the estimation of the models. Moreover, by not excluding them the model building exercise can become unnecessarily complex.
Undergraduates
According to DEET (1995a) there are two categories of undergraduates-Bachelor and Other Undergraduates. Six courses of varying length are included under these two categories. We have defined undergraduates as those students enrolled for the following three courses:
These three courses generally take three to four years of full-time study and tend to form a homogeneous group among the undergraduate courses. They comprised over 95 percent of all undergraduate enrolment in 1993 and 1994. A number of diploma courses have been converted to Bachelor courses in recent years, most notably in the nursing and teaching area.
The data on undergraduates is examined by four characteristics:
The number of years of enrolment has six categories-one, two, three, four, five and six or more.
Eleven broad fields of study are considered. These are:
These fields are the same as those defined by DEET (1995a) with two exceptions. First, Education has been split into two fields of study-initial training and the other. The initial training part of education is of importance in its own right because a high proportion of the supply of new graduate teachers come from this category. This field will be referred to as Education (I) and the rest of education as Education (O). Secondly, Medicine (not medical science) and Dentistry (not dental therapy) has been taken out of Health and combined with Veterinary Science to form one field which from now on will be referred to as Medicine. Combining Medicine, Dentistry and Veterinary Science into one field makes statistical sense because all three courses generally take about five years to complete.
As an initial analysis, tables were constructed to show:
This was done for each broad field of study. Moreover, the gender balance in each group defined by age and field of study, and number of years of enrolment and field of study, was investigated. In general, there was no significant difference between 1993 and 1994 in the age profiles, and the distribution by number of years of enrolment. Thus, only the analysis for students enrolled in 1994 is reported.
The age profile of students enrolled in undergraduate courses in 1994 is given in Table 3.1. The Australian category does not include students who are full fee-paying and from overseas. Only Australian students are included in the data for each broad field of study. Overall 34 percent of male students are under 20 years of age compared to 37 percent of female students. Similar percentages in the 20-24 age group are 41 and 35, and in the over 24 age group 26 and 28, for male and female students, respectively.
| Age in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female |
| Under 18 | 7 | 8 | 7 | 9 | 6 | 9 | 6 | 8 | 6 | 8 | 7 | 9 | 5 | 9 |
| 18 | 13 | 14 | 14 | 15 | 13 | 16 | 13 | 16 | 12 | 14 | 14 | 15 | 12 | 16 |
| 19 | 14 | 15 | 14 | 15 | 16 | 17 | 14 | 17 | 13 | 14 | 14 | 16 | 15 | 18 |
| 20 | 13 | 13 | 13 | 13 | 13 | 14 | 13 | 13 | 11 | 11 | 12 | 13 | 15 | 16 |
| 21 | 11 | 9 | 11 | 9 | 11 | 10 | 11 | 11 | 9 | 7 | 10 | 9 | 12 | 11 |
| 22 | 8 | 6 | 7 | 6 | 7 | 5 | 9 | 10 | 6 | 5 | 6 | 6 | 7 | 5 |
| 23 | 5 | 4 | 5 | 4 | 4 | 3 | 7 | 7 | 5 | 3 | 4 | 4 | 4 | 3 |
| 24 | 4 | 3 | 3 | 3 | 2 | 2 | 4 | 5 | 4 | 2 | 4 | 3 | 3 | 2 |
| 25-29 | 11 | 9 | 10 | 9 | 10 | 7 | 11 | 7 | 12 | 9 | 12 | 10 | 9 | 6 |
| 30-34 | 7 | 6 | 7 | 7 | 7 | 6 | 5 | 4 | 8 | 7 | 8 | 6 | 6 | 5 |
| Over 34 | 8 | 12 | 9 | 13 | 11 | 9 | 5 | 3 | 15 | 19 | 9 | 8 | 12 | 9 |
| Total | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 |
| Under 20 | 34 | 37 | 36 | 38 | 34 | 43 | 34 | 40 | 31 | 36 | 35 | 41 | 32 | 43 |
| 20-24 | 41 | 35 | 39 | 34 | 37 | 35 | 44 | 46 | 34 | 29 | 37 | 35 | 41 | 37 |
| Over 24 | 26 | 28 | 25 | 28 | 28 | 22 | 22 | 14 | 35 | 35 | 29 | 24 | 27 | 20 |
| Age in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | ||
Under 18 | ||||||||||||||
18 | ||||||||||||||
Total | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | ||
| Under 20 | 2 | 3 | 43 | 51 | 34 | 36 | 24 | 30 | 32 | 34 | 41 | 50 | ||
| 20-24 | 22 | 34 | 45 | 43 | 37 | 30 | 42 | 44 | 57 | 55 | 39 | 36 | ||
| Over 24 | 76 | 64 | 13 | 6 | 30 | 34 | 34 | 26 | 10 | 10 | 20 | 15 | ||
| Age in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
0 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 0 | 1 | |
The age profile of students vary a great deal by sex and also across fields of study. Except for the Education (O) field of study, in each field of study the proportion of females who are under 20 is higher than the proportion of males. For example, 43 percent of the females studying Education (I) are under 20 years old, but only 32 percent of the males studying Education (I) are in this age group. On the other hand in each field of study, except Medicine, the proportion of males who are 25 to 29 is higher than the proportion of females. In Medicine the proportion of males and females is the same in this age group.
Table 3.2 shows the age profile of students who completed courses in 1993. Most students tend to have finished the course when they were between 20 and 24 years of age. These data also have a pattern of variation which closely resembles that observed for the enrolment data.
The percentage breakdown by gender for students enrolled in 1994 is given in Table 3.3. For example, in Arts 25 percent of the under 18 age group is male and 75 percent is female. The aggregate figures indicate females outnumber males significantly in the younger age groups (20 years and younger) and the older age groups (30 years and over). However, this pattern is not uniform across all fields of study. Females seem to dominate across all age groups in Arts, Education (I) and Health, while male show dominance in Architecture and Engineering. In Science and Business males and females are in almost equal proportions in the younger age groups, but males are in higher proportions in the older age groups. Females are more numerous or equally as numerous as males in all age groups, except the over 24, in Law. In general, in Medicine there is a balance in the gender composition, except in the age groups between 21 and 29 when males are in higher numbers. Table 3.4 gives the gender balance in course completions. Once again, the pattern of variation is a reflection of that observed for the enrolment data..
Table 3.5 shows the variation in the number of years of enrolment (time in the system) for students by gender and field of study for 1994. For example, 31 percent of all female students in Architecture are in their first year of enrolment. On comparing with the data for 1993, which is not included in this report, 1994 data shows a significant increase in the percentage of both male and female students who are in the fourth or higher year of enrolment in Education (I), thus reflecting a reduced number commencing this field of study in 1994. Some differences between fields of study reflect the variation in the course length; for example, Engineering and Medicine are courses of longer duration, and thus, a higher proportion of students are in the fourth and fifth year of enrolment in these fields of study. Overall a female is less likely to be in her fourth or higher year of enrolment than a male is.
| Age in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female |
| Age in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | ||
The variation in the time to complete a course is given in Table 3.6. Overall 38 percent of females took three years to complete a course compared to 29 percent of males who took the same time. In general, a higher proportion of males took four or more years to complete their course than females across all fields of study. In particular, in Health and Education (I) the difference in this proportion between the two gender groups is quite large.
Table 3.7 shows the gender balance of undergraduate students enrolled in 1994 at each year of enrolment. For example, of all Arts students in their first year of enrolment 31 percent are male and 69 percent female. In general, at the fifth and higher year of enrolment males outnumber females, except in Arts, Education (I), Education (O) and Health.
The gender balance of undergraduate students who completed a course in 1993 is given in Table 3.8. For example, 34 percent of those students who completed a course in three years were male and 66 percent were female. The table shows that, in general, females make up a significantly larger proportion of the students who finish the course in four years or less, while males make up a higher proportion of those who take five or more years. However, there is considerable variation from this across fields of study. For example, males heavily outnumber females in Agriculture, Architecture and Engineering at all levels of completion times, while females outnumber males in Health.
Postgraduates
Three groups of postgraduate students are analysed. These are:
Unlike the analysis for undergraduate students postgraduates are not analysed by field of study. because there are unlikely to be enough students in each cell of the input-output matrix for reliable estimation of the model parameters. Further work in this area may be undertaken at a later date.
| Time in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female |
| Time in Years | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | ||
| Completion | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female |
| Completion | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male | Female | ||
The age profile of male and female students enrolled for each level of course for 1994 is given in Table 3.9. For example, it shows that 41 percent of all male Research students were under 31 years old, 33 percent were between 31 and 39 while 25 percent were over 39. The age profile for male Other Postgraduate students is almost identical. The age profile for male Master's by Coursework students show a higher proportion of them in the 31 to 39 age group then in the under 31 group.
Thirty-two percent of female Research students are over 39 years old, substantially higher than the corresponding figure for males. The age profile for females at both the Research and Master's by Coursework level is quite similar. However, the age profile for Other Postgraduate level students is substantially different. Students at this level tend to be younger with 47 percent of them under 31 years old.
Table 3.10 shows the age profile of students who completed postgraduate level courses in 1993. Apart from the fact that students are older, especially the Research level students, the profiles are a reflection of the pattern for the enrolment data
| Age in Years | ||||||
| Under 23 | ||||||
| 23-24 | ||||||
| 25-26 | ||||||
| 27-28 | ||||||
| 29-30 | ||||||
| 31-32 | ||||||
| 33-34 | ||||||
| 35-39 | ||||||
| 40-44 | ||||||
| Over 45 | ||||||
| Total | ||||||
| Under 31 | ||||||
| 31-39 | ||||||
| Over 39 | ||||||
The gender balance of postgraduate students enrolled in 1994 is given in Table 3.11. There are more male Research students than females in all age categories. At the Master's level there are more females than males in the under 23 and over 45 age groups, while in all other groups there are either more males or there is more or less a gender balance. Females outnumber males in all age groups at the Other Postgraduate level courses with the difference quite large in the younger age groups.
| Age in Years | ||||||
| Under 23 | ||||||
| 23-24 | ||||||
| 25-26 | ||||||
| 27-28 | ||||||
| 29-30 | ||||||
| 31-32 | ||||||
| 33-34 | ||||||
| 35-39 | ||||||
| 40-44 | ||||||
| Over 45 | ||||||
| Total | ||||||
Table 3.12 shows the gender balance of postgraduates who completed a course in 1993. In general, the pattern in this table is not too dissimilar to that for the enrolment data in Table 3.11. One difference is that for the Research and Master's by Coursework level courses the proportion of females completing is smaller than the corresponding proportion enrolled in each age category apart from the under 23 age.
Table 3.13 shows the variation in the time in the system for students enrolled in postgraduate courses in 1994. There is little difference in the pattern of variation between males and females within each level of course.
The distribution of time to complete a course for those who completed a course in 1993 is given in Table 3.14. It shows, for example, 71 percent of the males who completed Research degrees took four or more years to do so. The variation in time to complete a course is similar for males and females for each level of course.
Table 3.15 shows the gender distribution of students who were enrolled in a course in 1994. For example, 58 percent of Research students in their first year of enrolment were male, while the corresponding figure at the Other Postgraduate level is only 41 percent. In general, males tend to dominate at all stages of enrolment at the research level, while females tend to do the same at the Other Postgraduate level. The gender balance is less uneven at the Master's by Coursework level than for the other two levels.
Finally, the gender breakdown of those who completed a course in 1993 is given in Table 3.16. Males outnumber females at the Research and Master's by Coursework levels while the reverse is true at the Other Postgraduate level.
The data on school enrolment are required to estimate grade progression ratios, which in turn are used for projecting the number of students finishing Year 12 of secondary school. The school enrolment data for Australia by age, gender and grade for 1992 to 1994 was obtained from ABS (1993; 1994a; 1995a). For the purposes of making projections of higher education commencements up to the year 2001, data are required only on pupils in Year 6 or above in 1994. Pupils in Year 6 in 1994 will reach Year 12 in the year 2000. The ABS data had to be modified in a number of ways to make it consistent with the DEET higher education data. These modifications are described below.
Year 7 Aggregation
The ABS data contains two separate entries for Year 7 pupils. This is to distinguish the two structures of primary and secondary education that exist in Australia. In New South Wales, Victoria, the Australian Capital Territory and Tasmania Year 7 is included as part of the secondary education system, while in South Australia, Western Australia, Queensland and the Northern Territory it is part of the primary education system. For the purposes of this study the two entries for Year 7 were added to create just a single entry.
Ungraded Pupils
A number of pupils, both at the primary and secondary level, are ungraded. Some of these are in special education. The number of males who are ungraded is about 1.4 percent of the total number for both 1993 and 1994. The corresponding figure for females is substantially lower at about just 1 percent. In this study we distribute the ungraded pupils in a particular age group proportionally into grades using the shares of each grade in that age group as the proportion. An alternative procedure omitting these students could be considered.
Age Reference Date
The age reference date for the ABS data is first of July. However, for the DEET data on course enrolment and completions for higher education, this date is 31 December. In order to have a consistent definition of age the school enrolment data were adjusted. The adjustment process involved moving half the pupils in each age category into the next one.
The enrolment data, after it was adjusted as described above - for males, females and persons for 1994 - is given in Table 3.17. The percentage of students who are female in Years 6 to 12 is slightly lower than those who are male: there were 49.4 percent female students in 1994. However, female students made up 52.1 percent of the Year 12 population.
The comparison of the school enrolment figures for 1993 and 1994 show an increase in the number of students in Years 6, 7 and 8 in 1994, but a decline for Years 10, 11 and 12. Consequently we can expect a decline in Year 12 numbers for the next three years and then the numbers to pick up assuming constant retention rates.
Grade Progression and Retention Rates
The grade progression rate is simply the proportion of students from one grade level who progress to the next level in the following year. The retention rate is defined as the proportion of the Year 6 cohort students who progress on to Year 12. The overall retention rate, that is, the proportion of the Preparatory grade cohort students who progress to Year 12, would not be much different to the retention rate as almost all Preparatory students are expected to progress to Year 6. Since only stock data on student enrolment is available it is only possible to estimate these rates by the net grade progressions.
Table 3.18 gives the net grade progression rates and estimates of the retention rate. The difference in these rates for males and females are significant. Female progression rates are higher than that for males. Using the 1992 and 1993 stock data on student enrolment the retention rate for males and females is estimated to be 71 and 81 percent, respectively. However, these figures decline to 67 and 78 percent, respectively, when the 1993 and 1994 stock data are used.
Table 3.17 Full-time Students in Year 6 to 12, by Age and Grade, All Schools, 1994
| Age at 31.12.1994 | Year 6 | Year 7 | Year 8 | Year 9 | Year 10 | Year 11 | Year 12 | Total |
| 9 | 23 | 1 | 0 | 0 | 0 | 0 | 0 | 24 |
| 10 | 7784 | 37 | 4 | 0 | 0 | 0 | 0 | 7825 |
| 11 | 55017 | 7501 | 36 | 2 | 0 | 0 | 0 | 62555 |
| 12 | 58448 | 53689 | 7589 | 26 | 1 | 0 | 1 | 119755 |
| 13 | 11459 | 57349 | 53065 | 7414 | 30 | 1 | 1 | 129319 |
| 14 | 277 | 11451 | 56225 | 51267 | 7654 | 41 | 1 | 126916 |
| 15 | 15 | 350 | 11097 | 53955 | 48621 | 6953 | 43 | 121033 |
| 16 | 13 | 27 | 418 | 10574 | 50811 | 40253 | 6126 | 108223 |
| 17 | 10 | 12 | 42 | 563 | 10617 | 41846 | 33504 | 86593 |
| 18 | 2 | 11 | 5 | 80 | 976 | 9739 | 35136 | 45948 |
| 19 | 0 | 4 | 1 | 15 | 233 | 1654 | 9090 | 10996 |
| Over 19 | 2 | 5 | 8 | 15 | 323 | 1611 | 4013 | 5977 |
| Total | 133051 | 130437 | 128489 | 123910 | 119266 | 102098 | 87914 | 825166 |
| 9 | 22 | 1 | 0 | 0 | 0 | 0 | 0 | 23 |
| 10 | 8202 | 17 | 3 | 0 | 0 | 0 | 0 | 8222 |
| 11 | 54877 | 8007 | 46 | 2 | 0 | 0 | 0 | 62931 |
| 12 | 54508 | 53979 | 8182 | 36 | 1 | 0 | 0 | 116706 |
| 13 | 7960 | 53680 | 53226 | 8203 | 34 | 1 | 0 | 123103 |
| 14 | 160 | 7878 | 52283 | 51466 | 8535 | 48 | 0 | 120370 |
| 15 | 13 | 199 | 7446 | 50205 | 50064 | 8038 | 43 | 116008 |
| 16 | 7 | 18 | 272 | 7229 | 48406 | 43720 | 7472 | 107124 |
| 17 | 4 | 10 | 27 | 367 | 7383 | 42349 | 38902 | 89042 |
| 18 | 0 | 9 | 6 | 59 | 674 | 7528 | 37880 | 46154 |
| 19 | 1 | 3 | 3 | 21 | 185 | 1213 | 7450 | 8875 |
| Over 19 | 2 | 2 | 13 | 30 | 292 | 1581 | 3754 | 5673 |
| Total | 125754 | 123804 | 121506 | 117617 | 115574 | 104477 | 95499 | 804231 |
| 9 | 45 | 1 | 0 | 0 | 0 | 0 | 0 | 46 |
| 10 | 15986 | 54 | 7 | 0 | 0 | 0 | 0 | 16047 |
| 11 | 109893 | 15508 | 82 | 3 | 0 | 0 | 0 | 125487 |
| 12 | 112956 | 107668 | 15772 | 62 | 2 | 0 | 1 | 236460 |
| 13 | 19420 | 111030 | 106291 | 15617 | 64 | 2 | 1 | 252423 |
| 14 | 437 | 19329 | 108508 | 102734 | 16189 | 90 | 1 | 247286 |
| 15 | 28 | 549 | 18542 | 104159 | 98685 | 14991 | 86 | 237041 |
| 16 | 20 | 45 | 690 | 17803 | 99217 | 83973 | 13598 | 215347 |
| 17 | 14 | 22 | 69 | 930 | 18000 | 84195 | 72405 | 175635 |
| 18 | 2 | 20 | 10 | 139 | 1650 | 17267 | 73016 | 92103 |
| 19 | 1 | 7 | 3 | 37 | 418 | 2867 | 16540 | 19871 |
| Over 19 | 4 | 8 | 21 | 44 | 615 | 3192 | 7766 | 11650 |
| Total | 258805 | 254241 | 249995 | 241527 | 234839 | 206576 | 183413 | 1629397 |
Table 3.18 Estimates of Grade Progression and Retention Rates, All Schools
Grade Progression Rate | ||||||||
Year 12 Students Projections
On the basis of the grade progression ratios, estimated from the school enrolment data for 1993 and 1994, projections of student numbers in Year 12 by age and gender were made for the years 1995 to 2000. These are given in Table 3.19. The actual numbers for 1992 to 1994 are also included. Figure 3.1 shows time series plots of these projections. The number of Year 12 students is expected to decline until about 1997 and then slowly increase until the year 2000. However, this pattern of variation is not uniform across all age groups and gender. The over 19 female numbers are expected to continue to decline right up to the year 2000. In the under 18 and 18 year age group more female Year 12 students are projected in each year than male students, while in the 19 and over 19 age groups more male than female students are projected.
Table 3.19 Projection of Year 12 Enrolment, by Age and Sex, All Schools, 1995-2000
Figure 3.1 Projections of Year 12 Students, by Age and Sex, 1995 to 2000
The population estimates for Australia by age and gender for 1993 and 1994 were obtained from ABS (1995b) and the projections for the years 1995 to 2041 from ABS (1994b). The age reference date for these data is 30th June. The adjustment process, similar to that used for the school enrolment data, was also used on these data in order to have the date at which a student's age is calculated consistent with that used in DEET's files on higher education enrolment and completions.
Figure 3.2 and Figure 3.3 show the variation over time and difference between male and female population by various age groups of importance for this study. The plots reflect the decline in births in the 1970s and the consequent decline in the number of 15 to 17-year-olds from 1992 to 1994. The numbers of 15 to 17 year olds are projected to increase steadily after 1994, reflecting a slight recovery in births from 1980. The effect of the low birth rate also shows up in the plots of all age groups up to 24. There is a slump in the number of 30 to 34-year-olds from 1994 until 1999, due to the decline in births around the mid 1960s.
An exploratory analysis of the higher education, schools and population data showed pattern of variation by gender, age, field of study and level of course. The data on population projections show the effect of low births which occurred in the 1970s. In the next chapter, input-output models are fitted to the data. The results of estimating these models are presented and discussed.