Introduction
Searches through the DSU archives were able to obtain turnout data for the three major Union democratic bodies - Cross Campus Ballots, Union General Meeting, and Council. The records go back to 1979 in sufficient detail as to be able to obtain this data, however election turnouts are currently unavailable for 1988/9, 1993/4 and 1995/6, and no UGM turnouts are available before 1983/4.
The size of the University was drawn from various sources, giving data for years 1981/2, 1991/2, 1996/7 and 2000/1 onwards. It is known (from a University-supplied graph) that the increase in size was approximately linear between these points, and so the other figures have been interpolated.
Council turnouts
For each year in question, the peak Council turnout was used. For most years, the minimum Council turnout recorded was between 50% and 66% of the peak turnout.
Council turnout has fallen in absolute terms from around 120 in 1980 to around 60 now. Over the same time period, the number of (potential) voting members of Council has increased from approximately 50 to approximately 100.
Percentage turnout has fallen far more noticeably, going from 2.5-3% to just over 0.5% over the same time period. The correlation is visibly strong, (R2 = 0.93) and, if Council percentage attendance continues to fall at the same rate, will lead to the disappearance of Council by 2006. While the rate of decline is likely to level off over the next few years, this substantial and consistent decline is still concerning.
Returning to the absolute figures suggests that, while the number of voting members of Council attending has remained almost constant over the previous 20 years, the number of non-voting attendees has fallen from almost the same number to almost none.
Union General Meetings
The turnout for Union General Meetings is far more erratic than that of Council. Therefore, two figures have been used - the first is a 'typical' turnout for that year, determined by looking at as many records of attendance as exist for that year and selecting one near the median. The second is the peak turnout for that year - the highest turnout known from attendance records.
Peak turnouts usually, but not always, matched the appearance of an 'important' issue on the agenda. In some cases the peak turnout and typical turnout are identical. This may be due to incomplete records, or simply consistent attendance throughout the year.
Peak turnouts, where they are not the same as typical turnouts, vary widely, from being about 1% above typical to 6% above typical. Generally, where the peak turnout has been above typical turnout, this has not fallen as a percentage over the years, and has actually increased in absolute terms, the quorate UGM on 23 November 2000 having the greatest ever absolute attendance of any Union Meeting. Those of us who were there are quite aware of this!
Typical turnouts have been falling in relative terms on an approximately exponential curve. There was a massive drop at the end of 1987/8 and 1991/2, and a similar one at the end of 1994/5.
Since 1995, UGM turnouts have steadily risen, although in 2001/2 dropped again. Reasons for this are uncertain, and it will take a few more years before it becomes apparent if this is a one-off event or part of a pattern.
Possible causes for these will be looked at later.
Cross-campus ballots
Unlike turnouts at meetings, turnouts in cross-campus ballots have not been the subject of a gradual and consistent decline. The figure used is that of the highest turnout in that year. This was usually the election of the DSU President (and other sabbatical officers if held at the same time), but in a few cases (1991/2 and 1996/7) a referendum held in the same year got a significantly higher turnout, on one occasion almost double the (not small) sabb election turnout.
Percentage turnouts jumped almost at random between 1979 and 1992 from year to year, the period from 1982 to 1985 being the most chaotic. Since 1994, percentage turnouts have remained within a narrow range of 15% to 25%.
Absolute turnouts over the same period have showed a steady rise. The introduction of electronic voting has improved turnouts, into the 20-30% range, with much increased absolute turnouts.
Analysis
Election turnout is very variable. One thing that does stand out is that referenda tend to get significantly higher turnouts than sabb elections - of 25% compared to 17% over tuition fees, and of 47% compared to 27% over a rent strike continuation ballot. The consistency of turnouts over the last few years has no obvious cause, but seems to make more sense than the wide variation before that.
It is worth noting, for example, that in 1984/5, the number of people voting in the sabbatical election was barely more than the peak UGM turnout. It is possible that improvements in communications technology in recent years have put a far better minimum on the numbers aware of the election, and it will be interesting to see what effects the introduction of electronic voting will have on this.
The decline in Council and UGM turnouts is of greater concern. While there are diseconomies of scale associated with a larger body, these should only cause a fall in relative turnout. For the fall in relative turnout to be so rapid that absolute turnout also falls probably has another cause.
The reasons often suggested of "student apathy", "bad image of DSU" and "JCR/DSU conflict" may in part be correct. However, a search through the archives reveals that these are not recently perceived problems. Records mentioning two of these exist far further back - see the links. The problem of "student apathy" is not mentioned in any archives yet found before 1992 (but this is likely to be due to incomplete searching), however, turnout variances between colleges suggest that it is a symptom of a problem, not a problem in and of itself.
The drops in UGM relative turnout do not, in general, appear to co-incide with events in DSU's history that could have created a bad impression. They do also not generally coincide with drops in Council or Election turnout. One possible exception is between 1991 and 1992, where a very large drop in UGM turnout followed a year in which the Union had unsuccessfully led a rent strike (in which a late fee was charged). Turnout did, however, rise again after this, until 1994/5. It is uncertain from the archives what happened at this point - the lack of a Steering Committee for most of calendar year 1995 may have had a lot to do with this - so whether there was a similarly unfortunate event is unknown.
In 2001/2, the UGM turnout dropped again. The drop was not, however, substantial, and the smaller meetings were those for which there was almost entirely no business to discuss. The peak turnout recorded this year was for a meeting of Senate. However, this was probably comparable.
Whether the trends in Council and UGM turnout will continue or not is not yet predictable.