7 Predicting positive search outcomes
The analysis conducted so far has relied on a descriptive examination of change in the relationship between positive outcome and protected characteristics (age, sex and ethnicity) since the introduction of the CoP. However, there are a range of other factors that might have impacted on the success of searches, including when, where and why it took place. In addition, it is impossible to tell whether there may have been other unobserved changes in police approaches to searching that occurred as a result of the introduction of the CoP which had an independent effect on improving the success rates. Therefore, it is important to examine what had the biggest influence on detection rates. This was done using ‘regression analysis’ – a technique that enables the influence of multiple factors on an outcome (i.e. a positive search) to be tested simultaneously. A specific aim of this analysis was to establish whether the introduction of the CoP had an effect on the positive detection rate which was over and above that of the other factors that are known to influence detection. The results of this analysis are presented below.
7.2 Factors included in the regression model
A set of potential ‘explanatory variables’ that were likely to have an influence on the outcome of the search were identified. Within the stop and search database there is a limited number of variables available and it is likely that certain factors influenced detection rates that cannot be accounted for here. Nevertheless, three sets of factors were considered, as detailed below: 
1. The protected characteristics of the person who was searched:
- Sex (reference category: Female)
- Age group (reference category: 16-17 years)
- Ethnic group (reference category: White)
2. Factors relating to the nature of the search itself:
- Day of the week (reference category: Saturday)
- Time of day (reference category: 6pm-midnight)
- Reason for search (reference category: Drugs)
- Division (reference category: Greater Glasgow)
3. The time period of the search:
Before or after the CoP was introduced (reference category: Before the CoP)
The results of a regression model are expressed in terms of odds. This means that a value greater than one suggests that there are increased odds of a positive search due to the presence of a particular variable, while a value less than one suggests that there are decreased odds of a positive search in the presence of a particular variable. A value of one would suggest that the particular variable has no effect on whether or not the search results in a positive outcome. As all variables are tested simultaneously, the odds for any one factor are interpreted as being true when all the other variables in the model are held constant at their average values. Confidence intervals at the 95% level are included in the models to determine significant differences between the values of any one variable (e.g. differences between age groups).
7.3 Results of the regression model
Looking first at the protected characteristics, the regression model found that the odds of a search being positive were not significantly different for males and females when all other factors in the model were taken into account. In other words, the fact that there were more searches of males than females did not appear to result in any inequality in terms of effectiveness.
There was a significant difference in whether or not a search was likely to be positive based on the age of the person being searched. Figure 18 shows that searches of those aged under 18 had lower odds of resulting in a positive outcome than those of older people, even when all the other factors in the model were taken into account. This suggests that, even though the number of searches has declined markedly for this group, it appears that the decision to search younger people may still apply a lower threshold of reasonable suspicion than for those who are older.
Figure 18: Regression model predicting a positive search by age, controlling for other factors
There was no significance difference in the likelihood of a positive detection between searches involving white people and those from Asian, Black African, Black Caribbean or multiple/mixed ethnic backgrounds. Even though they were relatively small numbers, Figure 19 shows that the odds of a positive search was significantly lower for those from other ethnic groups or from those whose ethnicity was unknown compared to white people, even when taking account of many other factors. In other words, there may also be a lower threshold of suspicion used in searching people from these categories. Further work needs to be done to establish the ethnicity of people included in the ‘unknown’ category.
Figure 19: Regression model predicting a positive search by ethnicity, controlling for other factors
Turning now to the factors relating to the search itself, there was some difference in the likelihood of a positive search based on day of the week and time of the day. Searches conducted on Thursday to Sunday had greater odds of resulting in a positive outcome than those that happened on Monday to Wednesday, although the differences were not extensive.
Furthermore, searches that happened in the evening (6pm to midnight) had greater odds of resulting in a detection than those that occurred in the afternoon (midday to 6pm), which were in turn more likely to result in a detection than those that occurred between midnight and 12pm. These findings suggest that officers are more successful in detecting items during periods that would typically be associated with criminal or anti-social behaviour, whereas at ‘quieter’ times of the day or on ‘less busy’ days of the week they may waste time and resources conducting less successful searches.
Searches for stolen property had greater odds of resulting in a detection than those for drugs, as shown in Figure 20, even though drug searches made up the vast majority of all searches overall. Whereas searches for offensive weapons were by far the least likely to result in a positive outcome compared to all other search types, even when taking account of other factors that determine success. This suggests that a lower threshold of suspicion is used for weapon searches.
Figure 20: Regression model predicting a positive search by reason for search, controlling for other factors
There were some distinct differences in positive detection rates between different police Divisions. Figure 21 shows that, during the period studied, holding all other factors constant, Lanarkshire and Ayrshire Divisions recorded the highest odds of a positive outcome compared with all other Divisions. Greater Glasgow, the North East and Dumfries and Galloway had the lowest odds of positive detection, even when a wide range of other factors had been taken into account. The majority of Divisions had a detection rate greater than that for Greater Glasgow. These findings indicate that, even when controlling for a variety of factors that impact on positive outcome, there are geographical differences that are not explained by the data analysed here. This is deserving of further examination in the 12 month review.
Figure 21: Regression model predicting a positive search by police Division, controlling for other factors
Finally, Figure 22 reveals that searches conducted in the six month period following the introduction of the Code of Practice had significantly greater odds of resulting in a positive detection than those in the equivalent six month period of the previous year. Indeed, the odds of a positive detection were increased by around 33% in the period after the CoP came into being compared to the prior year. This finding was highly significant, even when all the other factors that can impact on a positive detection were held constant. In other words, it appears that there were other changes during the period following the introduction of the CoP that impacted on improving detection rates that were not measured by the data used in this model. This could include greater application of the rules of suspicion or more careful use of engagement with individuals prior to proceeding to search. These issues would benefit from further investigation through qualitative research as part of the 12 month review.
Figure 22: Regression model predicting a positive search by time period, controlling for other factors
Regression analysis was conducted to examine the impact of three different aspects of searches on the odds of a positive outcome: the protected characteristics of the person who was searched; the factors relating to the search itself; and the time period of the search. A key aim of this analysis was to establish whether the introduction of the CoP had an effect on positive outcomes which was over and above that of other factors that influence detection. The results showed that positive outcomes varied significantly by age and, to a lesser extent, ethnicity but not by gender, when other factors were taken into account.
The success of searches was also influenced to some extent by the time of day and day of the week they were conducted in, with evidence that some periods of activity are more productive than others. Successful detection was found to be greater in the case of searches for stolen property than drugs, but searches for offensive weapons were far less successful than any other type of search. Even taking account of these other factors, there were considerable differences in the likelihood of a successful outcome based on the Division in which the search took place. Ayrshire and Lanarkshire stood out as especially successful, with around three times greater odds of a search being positive compared to Greater Glasgow. These continued geographical differences would benefit from further evaluation in the 12 month review.
Finally, searches that were conducted in the six month period following the introduction of the Code of Practice were around 33% more likely in terms of odds to result in a positive detection than those in the equivalent six month period of the previous year. This is despite the fact that all the other factors influencing a positive detection were taken into account. Therefore, it appears that there has been a real, measurable improvement in the likelihood of a positive search during the period following the introduction of the CoP. It cannot be determined from this interim review what other factors might have led to this improvement; therefore, this would benefit from further investigation through qualitative research as part of the 12 month review.
Recommendations for the 12 month review:
11. To examine generally how practice in relation to search and seizure has changed within police Divisions as a result of the introduction of the Code of Practice and why this has led to an increase in positive search rates.