5. SUMMARY AND DISCUSSION
5.1 views of Licensees
In total, 130 licensees were interviewed during the baseline and impact phases of the evaluation. Sixty-seven interviews were carried out with on sale staff, and 63 licensees working in shops gave their views. All areas of Fife were represented in both research phases. Also, almost all of the licensees interviewed at baseline had not yet been targeted by the test purchase pilot, whereas the 30 interviewees at impact had all been subject to a recent test purchase visit. Twenty of the interviewees at impact were based in shops or bars that had failed the initial test purchase visit.
At baseline, off sale respondents seemed to be more likely to report that they had knowledge of test purchasing and the alcohol test purchase pilot than was the case with their on-sale counterparts. Indeed, one-half of on sale licensees said that they had little knowledge of test purchasing at baseline. As would be expected, awareness of test purchasing and the pilot increased at impact after the premises concerned had been subject to an actual test purchase visit, although only 30 impact interviews were carried out with licensees. At baseline 62% of licensees stated that test purchasing was a very good idea, with only four individuals arguing that it was a bad idea. At impact the majority of licensees still perceived that test purchasing was at least quite a good idea.
The licensees were asked how likely it would be that their outlets would be targeted during the test purchase pilot. At baseline, about three-quarters of respondents thought it likely that a test purchase visit would take place in their bar or shop. At impact, all 30 of the licensees said that it was likely that they would be subject to a test purchase visit in the future as part of the pilot exercise.
Although the majority of respondents also reported that proof of age card schemes were a good idea, they did not appear to be viewed as positively as test purchasing by the licensees. For example, over one-quarter of licensees working in the on sale sector viewed such schemes negatively at baseline. Not surprisingly, almost all of the licensees interviewed as part of both research phases said that they would act for proof of age documentation if they suspected that the person attempting to purchase alcohol might be under the legal age.
About two-thirds of licensees at baseline and impact stated that they gave their staff at least quite a lot of training in relation to alcohol sales to young people. Levels of reported training seemed to be lower in the on sales sector. There was some evidence that training levels increased at impact, which might have been expected as two-thirds of the outlets had recently failed a test purchase test. Even licensees who said that they only gave a little training at impact tended to say that the staff were now much more vigilant and aware of the procedures they should be following.
At baseline the majority of licensees perceived that the pilot would have little impact on their own retail practice. However, a slight majority of respondents also went on to say that the pilot would affect the practice of other bar and shop staff in the area who were viewed as knowingly selling alcohol to those under the age of 18 years. This pattern was also observed during the tobacco pilot evaluation, with interviewees arguing that there were unscrupulous traders operating within the same neighbourhoods. At impact, though, most licensees reported that the pilot had resulted in changes being made to their premises or retail practice, which is again to be expected given that most of the respondents during this phase had failed the initial test purchase visit.
The most popular measure among licensed staff at baseline and impact to reduce sales of alcohol to those under the age of 18 years was for test purchasing to be utilised in combination with other procedures, in most cases a robust, nationally-accepted proof of age card scheme. One-quarter of respondents thought that a national proof of age card scheme alone was the best way forward. However, at baseline 58% of licensees called for test purchasing to be utilised along with a proof of age card scheme, increased surveillance and education of licensed trade staff. At impact, 15 of the licensees who represented outlets that had failed the test purchase visit wanted test purchasing and a national proof of age card scheme to be adopted. It is also worth noting, though, that there was very little support for either the status quo of for test purchasing to operate in isolation from other measures. A few respondents felt that the police could offer more support to the licensed trade, and that the underlying problem of adults buying alcohol for younger people would not be affected by the pilot. These general findings, though, support the views of traders who were interviewed during the tobacco test purchase pilot. In this project the most popular measure called for by the retailers was for test purchasing to be implemented in tandem with a national proof of age card scheme. Therefore, it would appear that the negative views in relation to test purchasing are held by a minority of licensed trade and retail staff.
At impact, in the sample of premises which sold alcohol to volunteers, a slight majority had junior members of staff at the point of sale during the test purchase visit. Most of the licensees also said that the test purchase visit was not obvious to staff, regardless of the outcome of the test purchase. However, it is important to note that the majority of licensees in premises which failed or passed the initial test visit perceived that it had been conducted at least quite fairly. In the instances when alcohol was sold to volunteers the view was expressed that staff had not been vigilant enough and had not followed procedures. However, a few interviewees who represented establishments that failed the test thought that the volunteers had looked over the age of 18 years and that the police had arguably acted in an overly heavy-handed and indiscrete manner when charging the staff members. Again, the majority of traders interviewed during the tobacco pilot shared the view that the test purchase visits had been conducted fairly.
5.2 VIEWS OF VOLUNTEERS AND PARENTS
Six volunteers and their parents were interviewed at baseline and impact. At impact they had either turned 17 years or were just about to become 17 years old, and therefore their time in the pilot was at an end. Three male and 3 female volunteers and their mothers participated in both research phases. All three police divisional areas of Fife were represented by the volunteers and parents. (At impact another group of five volunteers and parents were recruited for the study; their views will be represented in the outcome report.)
At the time of recruitment to the pilot exercise most of the respondents had little knowledge of test purchasing or, indeed, the pilot in Fife. The volunteers were mostly recruited via local schools and the police also carried out home visits to inform the volunteers and parents fully about the test purchase pilot and to brief volunteers in readiness for going out on actual visits to shops and bars.
At impact, all of the volunteers stated that they had been sold alcohol during test purchase attempts. In total, the six volunteers said that they had been sold alcohol on 17 occasions. They also emphasised that the briefing and training organised by the police had prepared them for every eventuality, and no suggested improvements for the briefing were given. The parents also gave positive views in relation to the briefing carried out by police.
The volunteers tended to say that they liked working with the police, raising awareness of the issue of underage alcohol sales when they were actually sold alcohol, the excitement of participating in actual visits and receiving payment for the work. Their parents echoed these views, as well as stressing that the volunteers thoroughly enjoyed the insight they had gained into police operations. On the other hand, test purchase visits were also viewed as being both nerve-racking and in contrast relatively tedious - the volunteers did not enjoy having to wait for police to charge staff and complete the paperwork after a sale had occurred. Again, these features of test purchase visits were all noted during the tobacco pilot. As the volunteers had enjoyed taking part in the pilot the only reason given for no longer being involved in visits was that the age restrictions were now applying that they had turned 17 years of age.
All of the volunteers indicated that they would be willing and would have no concerns about taking part in future test purchase exercises if they were of an appropriate age. A few modifications for future test purchase activities were suggested. These suggestions included the fact that the police officers should be less obtrusive during test purchase visits, more police should be employed on exercises so that volunteers would not have to wait for licensees to be charged and for test purchase visits to utilise volunteers who are not known in particular neighbourhoods (as indeed the volunteers acknowledged was the norm within this pilot). The parents were also willing to allow their sons and daughters to take part in future test purchase schemes. Indeed, most of the parents were unable to give any suggestions as to how the alcohol test purchase pilot might be improved if it were to be rolled out in other areas of Scotland, and they were very positive about the way the police had worked with the volunteers. Two respondents did have minor complaints about the communication from police officers, for example, in terms of officially informing volunteers that their involvement in the pilot was at an end. As has been pointed out, though, few negative insights into involvement in the pilot were given and it was clear that the parents and volunteers were favourably impressed by their participation in the pilot.
5.3 VIEWS OF POLICE RESPONDENTS
Four members of police staff were interviewed at impact. They were selected in order to reflect both strategic and operational viewpoints of the pilot, as well as to cover the different police divisional areas in Fife.
In the main the police respondents thought that the licensees had reacted very positively to the pilot, even if they had been caught selling alcohol to volunteers during visits. However, the interviewees had also received criticism of the pilot from a minority of licensees who had failed the test purchase visit. The licensee interviews also demonstrated that up to the point of the impact phase the negative views of test purchasing as an enforcement tool were expressed by only a small proportion of licensees.
The police interviewees were all very satisfied with the way the volunteers had participated in the pilot. The initial recruitment of volunteers, the briefing programme, the test purchase outings and the debriefings conducted after the visits were all perceived to have operated as smoothly as possible. Although it was said that a few volunteers had stopped participating in the pilot the interviewees were confident that this was due to factors outwith the control of the police. According to the interviewees, the main benefits the volunteers had gained were valuable work experience, an insight into both the work of the police force and the issue of alcohol sales to young people. All of the respondents reported that the volunteers they had worked with would be willing to take part in future test purchase activities, assuming that they were of an appropriate age.
The view of the police interviewees was that the pilot had been operating very successfully. This was thought to be due to the groundwork and planning that had gone into the pilot exercise from both Fife Police Force and members of the wider steering group. Any logistical problems, and few were mentioned, were thought to have been dealt with as the pilot progressed. The interviewees also viewed positively the fact that the pilot had received a better reception from the public and licensees than might have been predicted and the public were also informing the police if they suspected that any traders were selling alcohol to underage youths. However, the time-consuming nature of the work and the demands this made on the time of police officers were not viewed as positively by the respondents. Of course, the fact that this was a pilot exercise had an impact on the workload of the police, and it is unlikely that if test purchasing were to be adopted more widely other police forces would attempt blanket test purchase coverage across all licensed premises or would have to deal with the same level of bureaucracy.
All of the interviewees thought alcohol test purchasing should be adopted in the future, albeit in combination with a variety of other measures such as a national proof of age card scheme, increased surveillance and education of licensed trade staff. It was emphasised that such approaches would work best if a partnership approach could be developed between the police and licensees. A continuous test purchase strategy was called for, as opposed to a pilot exercise, although as has been mentioned above it was not thought to be necessary to provide blanket coverage of premises but to target future visits on surveillance operations and information received from the public. A national proof of age card scheme was viewed as helping licensed trade staff become accustomed to ask for the buyer's age, and test purchasing visits could then assess how successfully retailers and bar staff were implementing such a policy.
The police respondents had few additional suggestions for modifications to be made if the pilot were to be rolled out across other areas of Scotland. Suggested changes included the need for reduced bureaucracy and for dedicated police teams to carry out test purchases, in order that the need for the continual briefing and rebriefing of staff would decrease. One factor which the interviewees wanted addressed was for the issue of shop staff below the age of 18 years serving alcohol to young people to be clarified, and possibly for this practice to be made illegal, as this group was viewed as being reluctant to ask for proof of age from their peers. A number of licensees agreed with the view that younger members of staff were loathe to ask for the age of those buying alcohol. Similarly, the issue of older people acting as agents and purchasing alcohol for young people was thought to be something which future schemes had to consider. Finally, calls were made for a more corporate approach to be developed and for funding, possibly from the Scottish Executive, to be made available for police forces to assist them implement test purchasing schemes as a partnership approach with the licensed trade. It must be emphasised again that some of the suggested changes related to the peculiar demands of the pilot as it was operating in Fife, and it is likely that the bureaucracy and blanket targeting of premises would not occur if the pilot were to be adopted across Scotland.
5.4 TEST PURCHASE VISITS
At impact the first 600 data sheets completed by police staff who had taken part in test purchase visits were analysed. About two-thirds of the data sheets covered off sale premises and one-third on sale outlets. This varied by area, with Eastern Division being much more likely to target on sale premises in the earlier phases of the project. About 38% of the data sheets were submitted by Western Division.
Test purchase attempts did not occur on 27 occasions for a variety of reasons including:
- The outlet concerned no longer sold alcohol or had closed down
- The police officers decided that there was a possible risk for the safety of the volunteers ( e.g. gangs of youths at the door of shop or bar)
- The licensee recognised one of the police officers conducting the visit.
The vast majority of the remaining visits (83%, n=475) were first visits to shops and bars, with 78 premises being subject to a test purchase revisit at impact. The 'other' category (n=20 visits) consisted of premises that were reported as selling alcohol to young people as a result of intelligence received by the police, and were often revisits.
In total, 17% of all test purchase visits resulted in volunteers being sold alcohol. As would be expected the percentage of outlets that failed the first test purchase visit was slightly higher, with 19% of shops or bars selling alcohol to volunteers. During the first visits a slightly higher percentage of on sale premises sold alcohol to young people, although this increased level of sale did not reach statistical significance. Six shops sold alcohol to volunteers during the revisit after failing the first test, and three other shops failed the test when they were targeted as a result of intelligence received during the pilot. In total, when all types of visit are considered, 17.9% and 16.8% of on and off sale premises respectively failed the test purchase tests. There was, though, variation in the levels of sales occurring within the different off and on sale categories. For example, all off-licences passed the tests (n=16), whereas 25% of supermarkets failed the first test purchase visit.
The eastern area of Fife reported the lowest level of sales when compared with the other two areas (first test only: 12.6% vs 21.2%; P<0.05). It should also be noted that in eastern Fife no outlet sold alcohol to volunteers during test purchase revisits, although 11% and 7% of revisits in West and Central Fife respectively resulted in revisit test failures. The pattern of sales to volunteers varied greatly in each area by type of outlet. The on sale sector was much more likely to sell alcohol to volunteers in East and Central Fife, whereas off sale premises yielded more sales in the Western area. However, the different targeting strategies employed by the different police divisions might be impacting on these results, as might be the different gender profile of test purchasers, and therefore this pattern of sales might not still be observed at outcome.
Male volunteers were more likely to be sold alcohol during the first test purchase visit when compared with their female counterparts (23.2% vs 15.5%, P<0.05). It is important to note, though, that when revisit sales were taken into account this difference was no longer statistically significant. Also, the major reason for the difference observed between the sales achieved by the different sexes was the higher rate of sales to male volunteers in Central area (27.5% male vs 11.3% female; P<0.01).
At impact there was some evidence of a trend towards increasing sales to volunteers during first test purchase visits as time progressed. However, relatively few first visits included in the analysis had taken place in November 2006. If this trend is maintained at outcome it is likely that licensees were on their guard in the summer of 2006 after the first wave of publicity but were not as vigilant as the test purchase pilot progressed.
As was suggested by the result of the licensee interviews at impact, junior sales or bar staff were more likely to sell alcohol to volunteers when compared with more senior staff. In fact, they were twice as likely to fail the first test purchase test when compared with the licence holders themselves (P<0.05). The police were only able to inform the licensee directly that a sale had occurred in 37% of cases, presumably as the licensees were not at the premises at the time of sale. It is possible that junior staff were not always passing on the information that test purchasing had taken place, particularly if the relevant outlet passed the test, although the results of the impact phase (with an admittedly small sample of licensees) suggest that licensees were aware that their premises had been visited as part of the pilot. Nevertheless, if awareness of test purchasing or the pilot is still relatively low at outcome, when it is likely that the vast majority of premises will have been subject to a test purchase visit, it might suggest that police officers will have to contact licensees directly to inform them that their premises have been targeted as part of the pilot.
At impact, there was little variation noted in the level of test purchase failures occurring in premises which had 'over 18' or 'over 21' policies. Similarly, premises which had been interviewed by researchers prior to test purchase visits had a similar level of sales to those which were not part of the baseline interview phase.
Shop and bar staff who reported hearing the radio adverts about the alcohol pilot in Fife were much less likely to sell alcohol to volunteers when compared with their counterparts who were unaware of the radio features (14% vs 33%; P<0.001). However, about 23% of the staff members in premises in which initial visits took place were unsure as to whether they had heard the adverts or not. Also, it might be argued that those who fail the test are more likely to deny previous knowledge of the pilot, from whatever source. In total, over 55% of shop and bar staff targeted during the early phase of the pilot for a first visit said that they had heard the adverts, but this ranged from 44% of staff in east Fife to 61% of staff in central Fife (P<0.01).