This thesis examines how burglars select a target and carry out a crime. The four research questions addressed by the two studies conducted for this thesis are: 1. What are the processes used by burglars to select a target, break in, steal, and distribute the proceeds? 2. What are the crucial decision making cues used to select a target? 3. What is the impact of various cues, cue alternatives, cue order and combinations selected on target attractiveness in a controlled situation? 4. Does age or experience interact with the effect of any cues? The results furnish discussion and increase the understanding and prevention of break and enter (B&E). The research for this thesis was conducted in two studies. For Study One interviews were conducted with fifty persons who attend a methadone clinic. Participants were past heroin users who financed their drug use through the commission of break and enter. Semi-structured interviews were conducted that covered their drug use and criminal involvement. From these interviews a list of seventeen cues was developed - cue 1 (dog), cue 2 (lighting), cue 3 (alarm), cue 4 (occupancy - lights/tv/radio), cue 5 (occupancy - car in driveway), cue 6 (affluence), cue 7 (doors/windows), cue 8 (locks), cue 9 (garage), cue 10 (fence), cue 11 (garden), cue 12 (location), cue 13 (people in the street), cue 14 (neighbourhood watch), cue 15 (weather), cue 16 (inside information) and cue 17 (street type). Participants used these seventeen cues to determine the vulnerability of a target. For Study Two a computer program was developed, in which these seventeen cues were arranged in various combinations, across twenty case studies. The computer program allowed subjects to access as much information about a case study as they needed to make a decision about its attractiveness as a burglary target. A new sample of ninety-six burglars were asked to view the twenty case studies and give each target a rating from '0' (not a B&E opportunity) to '100' (a definite B&E opportunity). Over the twenty case studies, the subjects only accessed one third of the available information to make a decision. The lower the final rating for a ease study the fewer cues were selected. Subjects were quickly deterred if the first one or two selections revealed deterrent alternatives. In contrast, if the initial selections revealed attractive alternatives the subjects were hard to deter even if subsequent cue selections revealed only deterrent alternatives. Four cues - cue 1 (dog), cue 3 (alann), cue 13 (people in the street) and cue 16 (inside information) - accounted for 91.77% of all first selections. Six cues - cue 1 (dog), 3 (alarm), cue 4 (occupancy - lights/tv/radio), cue 5 (occupancy - car in driveway), cue 13 (people in the street) and cue 16 (inside information) - accounted for 67.8% of all selections made. Clearly these six cues are very important to offenders and they should be closely examined in any prevention initiative. Results revealed that on 282 Occasions subjects viewed only one cue then made their decision based on this one piece of information. The most common single cue was reliable inside information that there was a large amount of cash inside the house or when a good alarm was present. Decision trees were developed which graphically trace the selections of subjects and the ratings given after each selection. The trees showed that subjects reached different conclusions from the same case study because they could select different cues. The selection of different cues from the same case study led to great variation in subsequent cue selections. The decision trees confirmed the earlier finding that subjects are much harder to deter when the first one or two selections had attractive alternatives even if subsequent selections had deterrent alternatives. Results of linear regressions revealed that every cue was significant as predictor of final rating at least twice, however three cues - cue 3 (alarm), cue 12 (location) and cue 16 (inside information) - were significant as predictors ten or more times. The 96 subjects were divided into four groups on the basis of age (young and old) and experience (experienced and inexperienced). The young and inexperienced group used an average of 188.3 cues across the twenty case studies, whereas the older and experienced group used an average of 43.8 cues. Older and experienced subjects were harder to deter, compared to younger and inexperienced subjects. As experience increased fewer cues were needed to reach a decision. The results showed that the variation in final rating for each case study was explained by a few cues. For example, in case study 16 the Adjusted R Square with all seventeen factors entered was .945. With only six cues as predictors the Adjusted R Square reduced slightly to .939. This shows that although cues are mentioned in the literature and were selected by subjects in this study they were often ineffectual and did not assist in explaining the final rating. The two most effective prevention measures were the deterrent alternatives for cue 3 (alarm) and cue 4 (occupancy -lights/tv/radio). The two most influential attractive alternatives were for cue 12 (location; house is located on a corner block) and cue 16 (inside information; from a reliable source you are told there could be a large amount of cash kept in the house). Overall, the linear models with interactions showed that the inexperienced subjects' decision making was more volatile and fluctuated to a greater extent than the experienced subjects' decision making. When continually attractive infonnation was received the inexperienced subjects' ratings climbed higher than did the experienced subjects. When deterrent information was received the negative effect on the inexperienced subjects' ratings was greater than the effect on experienced subjects. Experience increases burglars' skills and abilities but it also improves their capacity to weigh up information in a more reasoned manner. The results revealed that experienced subjects have probably developed a level of skill to the extent that the deterrent alternatives for many cues have become ineffectual. The experienced subjects have developed strategies to overcome many deterrents. The decision making of the experienced subjects was clearly more sophisticated and considered. The main theoretical finding of this thesis is that research will only produce incomplete findings if it concentrates on place and situation to the neglect of the offender and the antecedents and attributes they bring to a crime. The influence of age and experience on decision making is of such consequence that it must be considered to maximise the prevention of crime. Age and experience have individual and combined influences on cue selection and interpretation.