ALEX C. MICHALOS and BRUNO D. ZUMBO

PUBLIC SERVICES AND THE QUALITY OF LIFE (Accepted 17 December, 1997) ABSTRACT. This is a report of the results of a survey of citizen beliefs and attitudes about public services and the quality of life in Prince George, British Columbia, Canada in the summer of 1997. Information is provided about the perceived frequency with which various services were used, the perceived satisfaction and value for tax dollars spent on the services as well as on levels of government officials generally, preferences for the provision of more or fewer services and for spending relatively more or less revenue on different services, views about user-fees, and views about smoke-free public places and the likely impact of bylaw changes on people’s behaviour. Using such information, we examined correlations among perceived satisfaction, perceived value for money, use, spending preferences and demand, and, using multiple regression analysis, explained 66% of the variance in life satisfaction scores, 57% of the variance in satisfaction with the quality of life scores and 37% of the variance in happiness scores. Applying LISREL 8.14, it was shown that a model in which our three global indicators were explained by 13 domain indicators was superior to a model in which the latter indicators were explained by the former, i.e., a Bottom-Up model was superior to a Top-Down model. A simple linear model was also used to explain 32%, 20% and 19%, respectively, of the variance in satisfaction with municipal, provincial and federal government officials.

1. INTRODUCTION

In this essay, we report on the results of a survey of citizen beliefs and attitudes about public services and the quality of life in Prince George, British Columbia, Canada in the summer of 1997. The aim of the investigation was to obtain information about the perceived frequency with which various services were used, the perceived satisfaction and value for tax dollars spent on the services as well as on levels of government officials generally, preferences for the provision of more or fewer services and for spending relatively more or less revenue on different services, views about user-fees, and views about smoke-free public places and the likely impact of bylaw Social Indicators Research 48: 125–156, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.

126

ALEX C. MICHALOS AND BRUNO D. ZUMBO

changes on people’s behaviour. Using such information, we examined correlations among perceived satisfaction, perceived value for money, use, demand, and, using multiple regression analysis, explained respondents’ overall satisfaction with the quality of their lives, satisfaction with life as a whole, happiness and satisfaction with government officials. Some comparisons made with a similar survey taken in 1994 (Michalos, 1996). The structure of the essay is as follows. In the next section (2), we describe our sampling technique and questionnaire. Then we move to a description of our sample (3) and descriptive statistics regarding public service use, demand, value for money and spending preferences (4). In section (5) we review descriptive statistics on user-fee preferences, smoking and smoke-free establishments, and satisfaction levels for a variety of public services and other aspects of life. Following this descriptive work, we do some correlational analyses among our variables for use, demand, value, satisfaction and spending (6), explain people’s satisfaction with their lives (7) and government officials (8), and offer some concluding remarks (9).

2. SAMPLING TECHNIQUE AND QUESTIONNAIRE

Two thousand 10-page questionnaires were mailed to a simple random sample of Prince George’s 34,000 households in June/July 1997. The first page contained a general question designed to measure the extent to which respondents or their family members used 18 different public services. The response categories were “not at all = 1 point”, “seldom = 2” and “often = 3”. The second and third pages listed 58 “things affecting people today”, including the previously mentioned public services plus more personal things like family relations, health, jobs and so on. The extent to which people were satisfied with these things was measured on a 7-point Likert scale with response categories ranging from “very dissatisfied = 1 point”, through “an even balance = 4”, to “very satisfied = 7”. Following these questions, there was a 7-point question designed to measure the extent to which people were happy with their lives as

PUBLIC SERVICES AND THE QUALITY OF LIFE

127

a whole, ranging from “very unhappy = 1 point”, through “an even balance = 4”, to “very happy = 7”. These were followed by a page containing a general question designed to measure the extent to which respondents felt that more or fewer of 26 services should be provided by the city. The response categories were “fewer = 1”, “same = 2” and “more = 3”. On the next page respondents were asked if they thought they were “getting good value for the money” spent on the 26 services listed on the previous page. Response categories were “poor value = 1”, “average value = 2” and “good value = 3”. These were followed by a question designed to measure the extent to which respondents would like to see “less = 1 point”, “the same = 2” or “more = 3” of their tax dollars spent on each of the same 26 services. Next came three open-ended questions asking respondents, first, to indicate “the most important service provided by the city”, and second, to indicate a service that “the city provides that it need not provide”. The third question was: “If you could change anything about this community to improve the quality of life here, what would that change be?” We then raised the question: “Do you think the city should consider raising more revenue from user-fees and less from property taxes?” The response options were simply “yes” or “no”. Following that, we asked respondents to tell us whether they thought they got poor or good value for their tax dollars from federal, provincial and municipal governments. The options for each sort of government were “poor value = 1”, “average value = 2” and “good value = 3”. The questionnaire ended with two pages of demographic questions, but there was also a single page added on as a separate instrument with 5 questions related to strengthening the city’s smoking regulations in public places.

3. SAMPLE CHARACTERISTICS

By the end of July, 715 (36%) useable questionnaires were returned, which form the working data-set for the survey. Respondents included 311 (44%) females and 404 (56%) males. Four hundred and forty-seven (63%) were now married and living with their

128

ALEX C. MICHALOS AND BRUNO D. ZUMBO

spouse. The average age was 44, with a range from 18–88. Five hundred and seventy-five (80%) owned or were in the process of buying their home. A full 231 (32%) had some university education, with 114 (16%) holding a university degree. The first language of 629 (88%) respondents was English. Four hundred and fortyseven (63%) were employed full-time and 84 (12%) part-time. Two hundred and thirty-two (33%) had total family incomes of from $40,000 to $69,999. According to the 1991 census, which is the last official full population count available from Statistics Canada, 49% of the city’s adult residents were female, 40% were married and living with their spouse, 68% owned or were in the process of buying their home and 87% had English as their first language. The best estimates we can make from Statistics Canada’s population projections for the broader region of our Local Health Area 57 in 1997 indicate that about 30% of our population is in the 1–19 age group, 35% in the 20–39 age group and 35% in 40 or older group. It is unfortunate that we cannot get any more recent or accurate data to compare to our data set, but that is how matters stand now. Broadly speaking, then, our working sample is a bit older, contains a few more men, more married people and more home owners than the population from which it was drawn. Compared to the last broad-based survey undertaken in Prince George (Michalos, 1996), this sample is more representative of the population as a whole. Since the bulk of the population of Local Health Area 57 is in Prince George, we may assume that roughly half of our sampled population is in the 20–39 age group and the other half is in the 40 and above group. Our sample split is about 40% in the younger group and 60% in the older group. It would be possible to weight the responses of the younger group in order to produce results based on a 50/50 split, but we suspect that the younger people who took the time to respond to our questionnaire are probably a bit different from those who did not respond. There is always the general problem, of course, of not knowing exactly how nonrespondents differ from respondents, but we did not think it would improve our understanding by making guesses about the younger group and not about the older group, or about both groups. However, we undertook several analyses of the two groups to identify statistically significant

129

PUBLIC SERVICES AND THE QUALITY OF LIFE

EXHIBIT 1 Percent using services within the last year No

Seldom

Often

Libraries Parks Playgrounds Social and recreational facilities for seniors Recreation activities for youth Recreation activities for adults Sports facilities Support for organized athletics Garbage collection Water supply Sewage treatment Facilities for music, theatre, and art Police protection Fire protection Public Transit Land use planning Enforcement of land use bylaws Animal control

13 9 37 80 54 35 22 44 16 24 35 29 57 81 60 81 82 73

34 36 30 13 23 41 35 28 10 8 10 45 37 15 26 16 15 23

52 55 33 8 24 24 43 28 74 68 56 26 7 4 14 3 3 5

Average

46

25

29

differences and some of the results of these analyses are discussed below in Section 5.

4. RESULTS: USE, DEMANDS, VALUE AND SPENDING

Exhibit 1 lists the percents of respondents who personally or whose families used or contacted agencies often. The most frequently used community services within the last year were the municipal garbage collection system (74%), water system (68%), sewage treatment system (56%), parks and recreation (55%) and libraries (52%).

130

ALEX C. MICHALOS AND BRUNO D. ZUMBO

The least frequently used or contacted services were those dealing with land use planning and land use bylaws (3%), fire protection (4%) and animal control (5%). Exhibit 2 lists the percents of respondents indicating that they wanted more or fewer services provided by the city. By a wide margin 88% wanted the city to provide more road maintenance and repair services, while 11% wanted no change and only 0.3% wanted fewer of such services. Curiously, the second highest demand for more services was focused on promotion of the city. Sixty percent wanted the city to provide more promotion of Prince George, while 34% wanted no change and only 2% wanted fewer of such services. In third place, 57% wanted the city to provide more recreation activities for youth, 30% wanted no change and only 0.6% wanted fewer. Perhaps the most striking, but not unexpected, feature of Exhibit 2 is the very low average percentages of people wanting fewer services. On average only 1.8% of respondents expressed a preference for the city to provide fewer services, with a range running from 0.3% to 8%. The latter figure was related to public performances, which is remarkable because the city spends relatively little supporting public performances, music and theatre. Exhibit 3 lists the percents of respondents indicating that they got poor, average or good value for tax dollars spent on the 26 public services. On average 15% thought they got poor value for tax dollars, 47% thought they got average value and 24% thought they got good value. Highest marks went to libraries, with 53% saying they get good value from them. Next to libraries, 45% said they got good value for money spent on parks and garbage collection, and 43% thought they got good value on fire protection. Lowest marks went to road maintenance and repair, with 64% asserting that they got poor value for money spent in this area. Thirty-two percent thought they got poor value in public transit and 31% thought they got poor value for money spent on sidewalks. Exhibit 4 lists the percents of respondents asserting that less, the same or more tax dollars should be spent on each of the 26 public services. By far more people preferred to see more spending on road maintenance and repair than on anything else. Seventy-nine percent wanted more tax money spent in this area. Next to that at some

131

PUBLIC SERVICES AND THE QUALITY OF LIFE

EXHIBIT 2 Percent wanting more or fewer services provided by the City Fewer Same More Libraries Parks Playgrounds Social and recreational facilities for seniors Recreation activities for youth Recreation activities for adults Sports facilities Support for organized athletics Garbage collection Water supply Sewage treatment Facilities for music, theatre, and art Support for public performances/music/theatre Police protection Fire protection Road maintenance and repair Snow control/removal Street lighting Storm drainage Sidewalks Public Transit Land use planning Enforcement of land use bylaws Animal control Landscaping of boulevards/public lands Promotion of the City

0.7 0.6 0.9 0.6 0.6 1 4 5 1 0.3 0.3 6 8 0.6 0.4 0.3 1 0.4 0.3 0.4 3 2 2 2 2 2

75 62 56 29 30 49 56 54 85 84 77 51 45 44 74 11 53 59 73 47 47 50 47 56 46 34

20 35 33 41 57 38 33 27 11 10 11 37 39 51 19 88 46 39 20 50 40 24 26 34 49 60

Average

1.8

54

36

distance, 48% wanted more money spent on recreation activities for youth. Forty-six percent wanted more money spent on police protection. Compared to the 1994 survey (Michalos, 1996), these are interesting results. In 1994, 82% wanted more money spent on hospitals and clinics, 58% wanted more spent on public health

132

ALEX C. MICHALOS AND BRUNO D. ZUMBO

EXHIBIT 3 Percent saying they get good or poor value for tax dollars Poor Average Good value value value Libraries Parks Playgrounds Social and recreational facilities for seniors Recreation activities for youth Recreation activities for adults Sports facilities Support for organized athletics Garbage collection Water supply Sewage treatment Facilities for music, theatre, and art Support public performances music/theatre Police protection Fire protection Road maintenance and repair Snow control/removal Street lighting Storm drainage Sidewalks Public Transit Land use planning Enforcement of land use bylaws Animal control Landscaping of boulevards/public lands Promotion of the City

3 3 5 11 17 10 7 8 6 6 4 13 15 12 3 64 26 18 10 31 32 14 14 22 20 23

38 46 47 35 44 52 47 46 44 43 44 49 47 52 48 28 47 58 59 54 43 47 45 52 51 54

53 45 31 9 14 17 34 21 45 42 37 22 17 30 43 6 26 20 20 11 11 6 7 12 22 13

Average

15

47

24

133

PUBLIC SERVICES AND THE QUALITY OF LIFE

EXHIBIT 4 Percent saying more or less municipal tax money should be spent Less Same More Libraries Parks Playgrounds Social and recreational facilities for seniors Recreation activities for youth Recreation activities for adults Sports facilities Support for organized athletics Garbage collection Water supply Sewage treatment Facilities for music, theatre, and art Support public performances of music/theatre Police protection Fire protection Road maintenance and repair Snow control/removal Street lighting Storm drainage Sidewalks Public Transit Land use planning Enforcement of land use bylaws Animal control Landscaping of boulevards/public lands Promotion of the City Average

4 2 2 3 2 6 9 11 2 2 1 14 15 2 1 1 2 1 0.7 1 6 4 5 3 3 5

71 65 62 36 35 56 58 55 84 80 76 49 48 47 69 17 54 64 75 55 51 55 52 63 54 45

20 27 26 33 48 23 24 20 8 11 10 27 27 46 23 79 41 30 13 38 31 2 15 22 37 41

4

57

28

service and 56% wanted more spent on police protection. At that time 46% wanted more spent on highways, streets and roads. On average 4% of respondents indicated that less municipal tax money should be spent on the 26 listed public services, while 28% preferred to have more spent. Fifty-seven percent wanted to keep

134

ALEX C. MICHALOS AND BRUNO D. ZUMBO

EXHIBIT 5 Percent saying they get poor, average or good value for their tax dollars

Spent by the Federal Government Provincial Government Municipal Government

Poor value

Average value

Good value

61 49 14

27 39 54

1.6 2.8 24

spending at the same rate. Greatest support for spending less tax money was reported for public performances of music and theatre, and facilities for such things. The figures were 15% and 14%, respectively. In response to the open-ended question about the most important service provided by the city, 226 (32%) said it was police and fire protection, and ambulance service; 142 (20%) said it was sewer and garbage collection; 138 (19%) said snow removal and water; 91 (13%) said street and road maintenance; and 23 (3%) said personal health and safety. Three hundred and ninety-five (55%) respondents gave no response to our question about which services the city need not provide. In fact, of the relatively few responses received, 79 (11%) said that all city services were needed. Fifty-four (8%) wanted some reduced funding for the arts, 33 (5%) wanted reduced funding for organized athletics and 23 (3%) wanted reduced funding for recreation and sports facilities. Exhibit 5 gives the percents of people indicating that they got poor, average or good value for their tax dollars from each of the three levels of government. Clearly, the closer the government, the better its rating, although no government got very enthusiastic support. Twenty-four percent asserted that they got good value from the municipal government, compared to only 2.8% from the provincial government and 1.6% from the federal government. On the other hand, 61% thought they got poor value from the federal government, 49% thought they got poor value from the provincial government, but only 14% thought they got poor value from the municipal government.

PUBLIC SERVICES AND THE QUALITY OF LIFE

135

5. USER-FEES, SMOKING RESTRICTIONS AND SATISFACTION

Regarding user-fees, 45% claimed that the city should and 37% claimed that the city should not raise more money from this source. Assuming the sampling error margins were about 4 percentage points for our survey results, the support for user-fees might be as low as 41% and the opposition might be as high as 41%. Since our sample had a disproportionate number of property owners in it, one might assume some bias in favour of user-fees. So, a conservative reading of our figures would conclude that there is roughly as much support as there is opposition to the idea of increasing the use of user-fees as a source of public revenue. Regarding the question of strengthening the city bylaw against smoking in public places, Exhibit 6 lists the percents of people saying that if certain areas were made 100% smoke free, they would visit them more often, the same amount, less often and never. The most interesting columns in the exhibit are the two Net Gain columns on the right side of the exhibit. For the first column, Net Gain was defined as the difference between the percent who would go more often and the percent who would go less often. With this definition, restaurants would gain the most from a totally smokefree environment. Because 39% of respondents said they would visit smoke-free restaurants more often than they currently visit restaurants and 10% said they would visit them less often, restaurants could expect a 29 percentage point gain in business resulting from strengthening the city’s bylaw. Reading down this Net Gain column, one finds that in every case the owner’s of the various establishments would experience some net benefit from strengthening the bylaw, with casinos and bingo halls having the smallest benefits. For the second column, Net Gain was defined as the percent who would go more often minus the percent who would go less often or never. Defining Net Gain this way, three kinds of establishments would gain and three would lose. Restaurant owners would still gain the most, with a 22 percentage point increase. However, casinos, bingo halls and billiard rooms would all suffer a net loss. Because the “Never” response category was ambiguous between people who might be thinking “I normally go but would never go again” versus “I never go anyhow”, we cannot be sure if in every case those who responded by checking off “Never” would repre-

136

ALEX C. MICHALOS AND BRUNO D. ZUMBO

EXHIBIT 6 Percent indicating they would visit establishments more often, the same, less often or never if the establishments became 100% smoke-free

a b

Establishment

More often

Same

Less often

Never

Net gaina

Net gainb

Restaurant Pub/Lounge Bowling alley Bingo hall Casino Billiard room

39 30 25 13 12 16

43 38 49 38 39 39

10 10 5 4 5 5

7 17 10 29 28 25

29 20 20 9 7 11

22 3 10 −20 −21 −14

Net gain = % more often minus % less often Net gain = % more often minus (% less often plus % never)

sent lost customers or not for any given establishment. However, we suspect (and it is only a suspicion) that most people would read the question as asking how they would probably change their behaviour given the new situation of no longer being able to smoke in the various establishments. If so, that would mean that most of those who checked “Never” would be counted as new lost customers for some establishments. Therefore, we think the second way of measuring net gains and net losses to businesses is the most accurate of the two ways. Since 78% of our respondents reported that they were not and only 22% reported that they were smokers, these results are not surprising. However, since our sample is fairly representative of the population of Prince George, it is likely that any other randomly drawn sample would confirm our results. Exhibit 7 lists the average scores for the most satisfying and dissatisfying aspects of respondent’s lives. (Appendix 1 contains a list of all the satisfaction/dissatisfaction items with comparisons between the 1994 and 1997 surveys.) The three most satisfying aspects of people’s lives were three kinds of interpersonal relations, namely, satisfaction with living partners (mean = 6.1), family relations and friendships (5.9). These were also the top three areas in 1994 (Michalos, 1996). The mean score for satisfaction with life as

137

PUBLIC SERVICES AND THE QUALITY OF LIFE

EXHIBIT 7 Mean scores for most satisfying and dissatisfying things Mean

N

Most satisfying Living partner Family relations Friendships Life as a whole Quality of life Overall happiness Personal health Self-esteem

6.1 5.9 5.9 5.8 5.6 5.6 5.6 5.6

623 696 702 702 697 643 704 701

Most Dissatisfying Road maintenance Drug abuse Vehicle accidents Prov. gov. officials Fed. gov. officials Sidewalks Public transit

2.7 2.7 2.9 3.4 3.5 3.8 3.9

710 672 684 681 679 701 664

a whole was 5.8, and for the overall happiness and quality of life it was 5.6. Highest levels of dissatisfaction were expressed for road maintenance and repair, and drug and alcohol abuse (2.7), and the number of motor vehicle accidents (2.9). In 1994, drug and alcohol abuse was most dissatisfying and the number of motor vehicle accidents was second. As indicated earlier, in order to better understand the differences between our relatively younger (19–39) and older (40 and above) respondents, we undertook several analyses. Appendix 2 contains a list of all the domains in which there was a statistically significant difference in the levels of reported satisfaction between the 19–39 age group and the 40 and above age group at the 95% level of confidence or better. Five aspects of the figures in that list merit comment.

138

ALEX C. MICHALOS AND BRUNO D. ZUMBO

First, the fact that there are 30 domains listed in the Appendix implies that 53% of the 59 compared satisfaction and happiness levels were different for the two groups. Second, for 30 of the 31 domains listed, the older group had higher levels of satisfaction than the younger group, which is a fairly typical finding from several surveys taken around the world (Michalos, 1991). Third, the only case in which the younger group had a higher average level of satisfaction (actually less dissatisfaction) than the older group was that with the number of motor vehicle accidents. The figures show that the relatively younger group, which suffers most of the motor vehicle accidents, is less troubled than the older group by the accidents. Presumably there is some relationship between the younger group’s relative lack of disturbance and their higher accident rate. Fourth, it is worth noting that the older group is relatively more satisfied than the younger group with both recreation facilities for seniors and elderly care services. Since relatively more of the older group would be affected by such facilities and services, one might have expected them to be relatively more critical than the younger group. Fifth and finally, it is remarkable that this list contains no global indicators (i.e., overall happiness, satisfaction with life as a whole or with the quality of life), no specific domains related to interpersonal relations (i.e., satisfaction with living partners, family relations, friendships) and no mention of satisfaction with self-esteem (which has considerable explanatory power for all our global indicators, as shown in Section 7). That is to say, for all of the most personal matters, there are no statistically significant differences between the younger and older groups in our sample. In response to our question about things to change to improve the quality of life in Prince George, 146 (20%) respondents did not give us any answer. Of the responses we got, 89 (13%) recommended improving air quality, 60 (8%) said clean up the downtown area, 55 (8%) said reduce crime, 37 (5%) said improve bus service and 28 (4%) wanted improvements in road maintenance and city landscaping. In 1994, the top three recommendations were to beautify the central city, reduce crime and eliminate air pollution. This certainly indicates a pretty stable agenda for action over a three year period.

PUBLIC SERVICES AND THE QUALITY OF LIFE

139

6. CORRELATIONS AMONG SCORES FOR SATISFACTION, VALUE, USE, SPENDING AND DEMAND

Exhibit 8 lists the results of correlating scores (Pearson productmoment correlations) for levels of satisfaction with 18 public services, perceived value for money, amount of use, spending preferences and levels of demand. Positive correlations among these things indicate that they are directly related and negative correlations indicate that they are inversely related. In fact, there are only four negative correlations out of 72 listed in the exhibit. The first thing to be noticed about these correlations is that on average (last row of the exhibit) they have considerable variation. The average correlation between levels of satisfaction with services and perceived value for money was 0.45, which tends to validate both measures insofar as one would suppose that if people are satisfied with a service they probably think they are getting good value for their money. Inspection of the first column reveals fairly strong relationships ranging from 0.29 to 0.57. Because the square of a correlation coefficient equals the percent of variation explained by either variable in the pair, the correlation of 0.45 tells us that 20% of the variation in satisfaction scores could be explained by perceived value for money scores, and vice-versa. The average correlation between people’s use of a service and their perceived value for money from the service was 0.05, indicating that (on average) the fact that people use a service a little or a lot is practically independent of their judgments about whether they get poor or good value for the tax money spent on the service. Inspection of the second column shows that half the relationships were not statistically significant (indicated by 0). The average correlation between use and spending scores was 0.24 (third column), indicating that the more or less one used a service the more or less willing one was to see relatively more tax dollars spent on it. The average correlation between use and satisfaction scores was 0.13 (fourth column), indicating that the more or less one used a service the more or less satisfied one was with it, and vice-versa. In terms of variance explained, the previous correlation yielded only 6% and this one yielded only 2%. So we are dealing with variables that have very little explanatory power.

140

ALEX C. MICHALOS AND BRUNO D. ZUMBO

EXHIBIT 8 Correlations among scores for satisfaction, use, spending preferences, wants, and value for money Items

Libraries Parks Playgrounds Rec seniors Rec youths Rec adults Sports facilities Organized athletics Garbage collection Water supply Sewage treatment Facilities for music, theatre, and the art Police Fire Public transportation Land use planning Enforce land use bylaws Animal control Average

Satisfac Use / Use / Use / Use / Spend / / value value spend satisfac wants wants 0.48 0.45 0.41 0.42 0.4 0.39 0.37 0.29 0.57 0.55 0.45 0.42

0.21 0.17 0 0.15 0.14 0 0 0 0.11 0.17 0.19 0

0.34 0.25 0.32 0.17 0.21 0.26 0.42 0.35 0.17 0.14 0.14 0.38

0.31 0.25 0.18 0.17 0.18 0.08 0.21 0.15 0.24 0.24 0.35 0.14

0.28 0.2 0.3 0.18 0.24 0.32 0 0.38 0.21 0.26 0.19 0.41

0.46 0.51 0.52 0.59 0.58 0.54 0 0.54 0.54 0.43 0.39 0.71

0.55 0.41 0.57 0.5 0.48 0.56 0.45

0 0 0 0 −0.1 −0.11 0.05

0.16 0.14 0.29 0.12 0.15 0.23 0.24

0 0.13 0 0 −0.1 −0.1 0.13

0.11 0.08 0.3 0.18 0.2 0.23 0.22

0.58 0.54 0.76 0.38 0.45 0.58 0.5

The average correlation between use and amount of service wanted was 0.22 (fifth column), and the strongest average correlation of all was between willingness to spend tax dollars and amount of service wanted, 0.50 (last column). This last correlation roughly tells us the extent to which people were putting their money where their mouths were, insofar as the more or less they wanted service the more or less they were willing to spend on it. Unfortunately our questions were not worded in such a way that they allow us to determine conclusively whether people were

PUBLIC SERVICES AND THE QUALITY OF LIFE

141

expressing a willingness to have taxes increased in order to purchase more services or whether they were merely expressing a willingness to have the allocation of their current tax dollars altered to purchase more of the services they wanted more of and less of the services they wanted less of. However, we think the wording probably suggested that we were interested in their views about reallocations. Besides, we know that there is considerable evidence from other national surveys in Canada indicating that most Canadians do not ever want to see the absolute value of their taxes increased; e.g., see Michalos (1988 and 1997). While most of the average correlations are modest, there are some robust individual correlations. For example, respondents’ demand for more public transportation services was positively correlated at 0.76 with their willingness to spend relatively more tax dollars on such services, and their demand for more facilities for music, theatre and the arts was positively correlated at 0.71 with their willingness to spend more on these services.

7. EXPLAINING LIFE SATISFACTION, HAPPINESS AND SATISFACTION WITH THE QUALITY OF LIFE FROM DOMAIN SATISFACTION, AND VICE-VERSA

Exhibit 9 provides some comparative figures resulting from regressing mean life satisfaction scores on mean scores for satisfaction with various domains of life. The simple linear model used to obtain these figures was one of the first models applied by social indicators researchers and it is still one of the most frequently used models (Michalos, 1991). Since the model merely explains overall life (global) satisfaction in terms of the (domain) satisfaction one gets from one’s job, family relations and so on, it is useless for explaining satisfaction in the generic sense. For the latter, one would require something other than satisfaction to do the explanatory work. For example, in multiple discrepancies theory satisfaction is posited as the effect of things like the perceived discrepancies between what one has and wants, and between what one has and others have (Michalos, 1985). Nevertheless, although the simple linear model is not deep, it does allow one to measure the impact of satisfaction with various domains of life on one’s perceived overall satisfaction

142

ALEX C. MICHALOS AND BRUNO D. ZUMBO

with life. Here we will also apply the model to explain happiness with life as a whole and satisfaction with the quality of one’s life. On average, for the seven samples represented in Exhibit 9, we were able to explain 55% of the variance in reported life satisfaction from some subset of the 14 predictor variables listed in the exhibit. Our best success came from the Prince George sample in 1997 (column G), at 64%. When all variables are standardized to have means of zero and standard deviations of one, the standardized regression coefficients (Betas) measure the percent of movement in the dependent variable when a predictor variable moves one full unit and every other predictor in the set is held constant. For example, under column G one finds that the Beta value for self-esteem is B = 0.57, which means that for every full unit increase in satisfaction with one’s own self-esteem, one’s life satisfaction increases half a unit. Inspection of the other figures in that column reveals that satisfaction in no other domain had as great an impact on overall life satisfaction for the 528 Prince George respondents. Reading across the row to the left from 0.57, one sees that satisfaction with one’s own self-esteem was also the strongest predictor of life satisfaction for the 430 Prince George residents in the 1994 sample (column F, B = 0.50) and for the sample of 296 University of Guelph students (column D), at B = 0.31. That Guelph sample was part of the global student sample whose results are listed in column E. Inspection of that column shows that satisfaction with one’s own self-esteem tied with satisfaction with one’s university education as the strongest predictors of life satisfaction for the 5107 students in the global sample. For the other three samples, satisfaction with one’s self-esteem was not as powerful a predictor as (had less explanatory power than) satisfaction with one or more other domains. For the 312 sampled members of the office, clerical and technical staff of the University of Guelph (column A), satisfaction with one’s family relations had the greatest explanatory power (B = 0.35). For the 273 sampled rural seniors (column B), satisfaction with their housing had the strongest predictive strength (B = 0.21) and for the 328 eastern northerners (column C), satisfaction with their financial security had the greatest explanatory power (B = 0.24).

143

PUBLIC SERVICES AND THE QUALITY OF LIFE

EXHIBIT 9 Comparison of life satisfaction regressions for university clerical staff, rural seniors, Eastern northern community, Guelph students, world sample of students, and Prince George residents in 1994 and 1997

Percent of variance explained in life satisfaction

1979 Aa

1981 B

57

49

1982 1984 1985/86 1994 1997 C D E F G 53

53

49

60

Predictors Satisfaction with: Standardized regression coefficients Health 0.11 0.18 0.17 0.12 0.11 b Financial security 0.15 −0.01 0.24 0.11 0.16 b Family relations 0.35 0.10 0.10 0.13 0.06 b Job 0.10 c b 0.09 b 0.08 Friendships 0.20 0.08 0.07 0.17 0.08 0.13 Housing −0.05 0.21 0.10 0.12 0.12 0.14 Area lived in b 0.01 0.13 b c b Recreation activity 0.08 0.08 0.05 0.12 0.13 b Religion c 0.13 0.07 b b b Self-esteem 0.13 0.17 0.19 0.31 0.19 0.50 Transportation 0.09 0.05 0.06 b 0.05 b Gov. services c 0.13 0.04 c c 0.13 Living partner c 0.06 c c 0.15 0.09 Education −0.03 c c 0.16 0.19 b

64

b 0.09 0.09 0.08 0.07 b b b b 0.57 b b 0.13 c

P < 0.01, a: A = Clerical staff (N = 312), B = Seniors (N = 273), C = East. northerners (N = 328), D = Guelph students (N = 296), E = world students (N = 5107), F = Prince George residents in 1994 (N = 430), G = Prince George residents in 1997 (N = 512). b: Significance level too low to enter equation. c: Not in questionnaire.

The figures just reviewed clearly illustrate the fact that different groups of people with different life circumstances, resources and constraints use different mixtures of ingredients to determine their life satisfaction. The explanatory model used here allowed the groups represented in Exhibit 9 to draw satisfaction from roughly

144

ALEX C. MICHALOS AND BRUNO D. ZUMBO

the same set of 14 domains. Analytically speaking, some groups used more and some used fewer of these domains to build their life satisfaction. The Prince George group used only six of the 14 possibilities, but not the same six in both years. In 1997, satisfaction with financial security and family relations had a statistically significant impact on overall life satisfaction, but no impact in 1994. In 1994, satisfaction with housing and government services had a significant impact on overall life satisfaction, but no impact in 1997. Exhibit 10 provides some comparative figures resulting from regressing mean happiness scores on mean scores for satisfaction with various domains of life. On average, for the seven samples represented in this exhibit, we were able to explain 35% of the variance in reported happiness with life as a whole from some subset of the 14 predictor variables listed in the exhibit. Our best success came from the clerical staff sample from the University of Guelph (column A), at 45%. The strongest explanatory variable for that sample was satisfaction with family relations, which had a Beta value of B = 0.38. Thus, figuratively speaking, for every fullstep increase in satisfaction with their family relations, the overall happiness of these respondents increased 38% of a step. As one moves across the columns of Exhibit 10, one finds that the strongest explanatory variable for the rural seniors group was satisfaction with their friendships (B = 0.23), satisfaction with their financial security was the strongest predictor for the eastern northern group (B = 0.21), satisfaction with self-esteem for the Guelph students (B = 0.26) and the 1997 Prince George sample (B = 0.27), and satisfaction with living partners for the world student sample (B = 0.18) and the 1994 Prince George sample (B = 0.22). As in the case of life satisfaction, different groups built up their overall happiness with different mixtures of ingredients and some groups drew upon more domains than others. Having examined the relationships between 13 domain satisfaction scores and the two global indicators, life satisfaction and happiness, with the former scores as predictors of the latter, it was a small step to reverse the direction of analysis. In fact, we added our third global indicator, satisfaction with the quality of one’s life, and we undertook two new analyses.

145

PUBLIC SERVICES AND THE QUALITY OF LIFE

EXHIBIT 10 Comparison of happiness regressions for university clerical staff, rural seniors, Eastern northern community, Guelph students, world sample of students and Prince George residents in 1994 and 1997 1979 Aa Percent of variance explained in life satisfaction

45

1981 1982 1984 1985/86 1994 1997 B C D E F G 32

36

39

28

27

Predictors Satisfaction with: Standardized regression coefficients Health 0.12 0.12 0.18 0.17 0.09 b Financial security 0.09 0.06 0.21 b 0.13 b Family relations 0.38 −0.03 0.09 0.14 0.06 b Job 0.03 c 0.09 0.18 b 0.10 Friendships 0.23 0.23 0.01 0.21 0.10 0.10 Housing 0.01 −0.01 0.10 b 0.07 b Area lived in b 0.01 0.05 b c b Recreation activity 0.03 0.04 0.05 b 0.07 0.13 Religion c 0.01 0.03 b b b Self-esteem 0.07 0.09 0.14 0.26 0.13 0.21 Transportation 0.05 0.05 b b b b Gov. services c 0.08 0.03 c c b Living partner c 0.30 c c 0.18 0.22 Education −0.03 c c b 0.10 b

38

b 0.14 0.14 b 0.09 b b b b 0.27 b 0.09 0.22 c

P < 0.01, a: A = Clerical staff (N = 312), B = Seniors (N = 273), C = East. Northerners (N = 328), D = Guelph students (N = 296), E = world students (N = 5036), F = Prince George residents (N = 412), G = Prince George residents in 1997 (N = 466). b: Significance level too low to enter equation. c: Not in questionnaire.

First, with LISREL 8.14 we applied a canonical correlation model to measure the linear relationship between the set of 13 domain satisfaction scores and the set of 3 global indicator scores. In effect, this procedure treats each of the two sets of scores as measures of a single dimension or factor, and the canonical correlation

146

ALEX C. MICHALOS AND BRUNO D. ZUMBO

coefficient for the pair is like an ordinary Pearson-product moment correlation coefficient. The canonical correlation for our two sets of scores was a very robust 0.92, suggesting that we probably have two alternative ways of measuring a single dimension or factor conceptualizable as quality of life. We believe this was the first time canonical correlation has been applied in quality of life research. Second, we applied LISREL 8.14 to compare the goodness of fit for two explanatory models, one in which the 13 domain satisfaction scores were used to explain the three global indicator scores (i.e., the so-called Bottom-Up analysis) and the other in which the three global indicator scores were used to explain the 13 domain satisfaction scores (i.e., the Top-Down analysis). Our dataset did not permit a Bi-Directional analysis (Mallard, Lance and Michalos, 1997). The results of these analyses indicated that the Bottom-Up model was superior to the Top-Down model, with the former having a goodness of fit index (GFI) of 0.97, compared to 0.80 for the latter. In the light of the results of our canonical correlation, it would appear that all one can conclude from this model-fitting exercise is that, for this dataset at least, a model in which many indicators are used to explain a few yields better fit scores than a model in which a few indicators are used to explain many. We believe this was also the first time LISREL has been used to fit models with three global indicators for quality of life and 13 domain satisfaction indicators. In order to be able to make rough comparisons among the seven groups considered in Exhibits 9 and 10, it was necessary to use a relatively small and consistent set of 14 predictors. However, in the interest of discovering some additional explanatory variables, more regressions were run testing many more (66) predictors that were available in the Prince George 1997 survey. As things turned out, in the presence of some new variables, the explanatory power of some of those in the older set was significantly reduced. Exhibit 11 gives the results of regressing the three dependent variables on the 66 predictors, which was accomplished in several steps using some zero-order correlation and repeated multiple regression analyses. We were able to explain 37% of the variance in reported overall happiness, 66% of the variance in reported satisfaction with life as a whole and 57% of the variance in reported satisfaction with the overall quality of life. Thus, using all the addi-

PUBLIC SERVICES AND THE QUALITY OF LIFE

147

tional potentially explanatory variables available to us, we were only able to increase our capacity to explain reported satisfaction with life as a whole by 2 percentage points and we actually lost a percentage point of explanatory power for happiness. Presumably the latter loss was the result of one or more variables suppressing the power of others. Perhaps the most interesting discovery from these regressions is the fact that only two predictors, satisfaction with one’s own selfesteem and satisfaction with local job opportunities, have roles to play in explaining all three dependent variables and the roles are different in each case. Regarding happiness and satisfaction with life as a whole, satisfaction with one’s own self-esteem is more influential than each of the other predictors. But in the context of the particular sets of variables involved in each case, there is a big difference in the relative amount of influence that the latter variable has on the former variables. Satisfaction with one’s own self-esteem (B = 0.24) has only a bit more influence than satisfaction with one’s living partner (B = 0.20) on happiness and nearly three times the influence of satisfaction with job opportunities (B = 0.09). Satisfaction with one’s own self-esteem (B = 0.56) has nearly four times more influence than satisfaction with one’s living partner (B = 0.15) on satisfaction with life as a whole and seven times more influence than satisfaction with job opportunities. Regarding satisfaction with the overall quality of life, satisfaction with one’s own self-esteem (B = 0.25) has half the influence of satisfaction with one’s standard of living (B = 0.50), while satisfaction with local job opportunities (B = 0.11) has only a fifth of the explanatory power of the leader. Reflecting on the list of variables predicting satisfaction with the quality of life in Prince George in 1997, it seems that the idea of “quality of life” has a fairly economic or materialistic connotation that is not shared by either the idea of “life satisfaction” or the idea of “happiness”. In particular, the idea of “quality of life” seems to be closely connected to the older concept of a “standard of living”. In the 1994 survey, satisfaction with one’s standard of living also had the greatest impact on satisfaction with the quality of life, but it did not dominate the set of predictors as it did in 1997. It also had a significant impact on life satisfaction in 1994.

148

ALEX C. MICHALOS AND BRUNO D. ZUMBO

EXHIBIT 11 Explanations of Happiness, Satisfaction with Life as a Whole and Satisfaction with the Quality of Life Happiness Predictors = Satisfaction with Family relations Living partner Financial security Self-esteem Job opportunities Weather Variance explained

Beta 0.16 0.20 0.14 0.24 0.09 0.12 37% (N = 501)

Life as a Whole Predictors = Satisfaction with Living partner Job Friendships Health Self-esteem Fed. gov. officials Geographic location Job opportunities Variance explained

Beta 0.15 0.07 0.09 0.07 0.56 0.05 0.09 0.08 66% (N = 528)

Quality of Life Predictors = Satisfaction with Overall living standard Self-esteem Job opportunities Geographic location Variance explained

Beta 0.50 0.25 0.11 0.13 57% (N = 640)

PUBLIC SERVICES AND THE QUALITY OF LIFE

149

8. EXPLAINING SATISFACTION WITH LOCAL, PROVINCIAL, FEDERAL GOVERNMENT OFFICIALS FROM SATISFACTION WITH PUBLIC SERVICES

Exhibit 12 gives the results of our attempts to explain respondents’ levels of satisfaction with their government officials on the basis of their levels of satisfaction with public services. Peoples’ impressions, beliefs and feelings about their government officials are certainly determined by a wide variety of things, e.g., news media reports about individual officials and policies, actual impacts of policies, programs and individual officials’ actions, imagined impacts, halo effects connected to stereotypes of public service workers and politicians, personal knowledge about particular officials, gossip and so on. So, we did not expect to be able to explain anywhere near one hundred percent of the variance in any levels of satisfaction with government officials from levels of satisfaction with public services. Nevertheless, what we found was very instructive. First, we found that our capacity to explain levels of satisfaction with government officials decreased as we moved the government farther from the people. We were able to explain 32%, 20% and 19% of the variance in reported satisfaction with municipal, provincial and federal government officials, respectively, on the basis of levels of satisfaction with public services. So, roughly 70 to 80% of the variation in people’s levels of satisfaction with their government officials must be explained by something other than the public services delivered by those officials. Clearly, those who are currently engaged in the search for key performance indicators for elected and unelected government officials will have to look somewhere beyond mere citizen satisfaction with services delivered. Second, our analyses found that respondents apparently did not have a clear idea of which level of government delivered which services. For example, satisfaction with libraries had a positive impact on satisfaction with Federal Government Officials, although there is no direct connection between the operation of our local library system and Federal Government Officials. Third, our reflections on the apparent confusion about which levels of government are responsible for which services led to the idea that in future surveys services should be clustered by the level

150

ALEX C. MICHALOS AND BRUNO D. ZUMBO

EXHIBIT 12 Explanations of Satisfaction with Local Government Officials, Provincial Government Officials, Federal Government Officials and All Government Officials Satisfaction with Local Government Officials Predictors = Satisfaction with Libraries Recreation activities for youth Snow control and removal Animal control Physical beauty of P.G. Hospitals and clinics Elderly care services Variance explained

Beta 0.12 0.11 0.11 0.08 0.10 0.23 0.19 32% (N = 457)

Satisfaction with Provincial Government Officials Predictors = Satisfaction with Hospitals and clinics Road maintenance Animal control Ambulance services Variance explained

Beta 0.25 0.16 0.08 0.17 20% (N = 542)

Satisfaction with Federal Government Officials Predictors = Satisfaction with Libraries Road maintenance Ambulance services Elderly care services Variance explained

Beta 0.14 0.21 0.20 0.15 19% (N = 460)

PUBLIC SERVICES AND THE QUALITY OF LIFE

151

EXHIBIT 12 Continued Satisfaction with All Government Officials Predictors = Satisfaction with Libraries Seniors’ recreation facilities Road maintenance Hospitals and clinics Ambulance services Elderly care services Variance explained

Beta 0.12 0.09 0.18 0.23 0.17 0.11 31% (N = 531)

of government delivering and/or paying for them. That way the survey questions would have some immediate educational value for respondents and would provide all of us with more reliable and valid answers. It also occurred to us to provide some sort of pie diagrams or charts indicating relative amounts of spending in order to give respondents an idea of how their tax dollars are currently spent before they answer questions about whether they would like to see relatively more or less spent here or there. 9. CONCLUSION

Using a simple random sample of 715 residents of Prince George, we explored people’s use of public services, their demand for more or fewer services, their perceived value received for tax dollars spent, their relative spending preferences for tax revenues, their satisfaction with public services and a wide variety of aspects of their community and their lives, and correlations among these things. A simple linear model was used to explain life satisfaction, happiness and satisfaction with the quality of respondents’ lives on the basis of satisfaction they received from public services and from specific domains of their lives, like job satisfaction and satisfaction with their family relations. This model was able to account for 66% of the variance in life satisfaction scores, 57% of the variance in

152

ALEX C. MICHALOS AND BRUNO D. ZUMBO

satisfaction with the quality of life scores and 37% of the variance in happiness scores. Applying LISREL 8.14, it was shown that a model in which our three global indicators were explained by 13 domain indicators was superior to a model in which the latter indicators were explained by the former, i.e., a Bottom-Up model was superior to a Top-Down model.A simple linear model was also used to explain 32%, 20% and 19%, respectively, of the variance in satisfaction with municipal, provincial and federal government officials.

ACKNOWLEDGEMENTS

The authors would like to express their appreciation to Alan Chabot, Toni Fletcher, Bill Kennedy, Wendy Marks and Shelley Rennick for the help they gave us in the development, administration and analysis of the survey.

REFERENCES Carmines, E. G. and R. A. Zeller: 1979, Reliability and Validity Assessment (Sage Publications, Beverly Hills). Mallard, A. G. C., C. E. Lance and A. C. Michalos: 1997, ‘Culture as a moderator of overall life satisfaction – life facet satisfaction relationships’, Social Indicators Research 40, pp. 259–284. Michalos, A. C.: 1980, ‘Satisfaction and happiness’, Social Indicators Research 8, pp. 385–422. Michalos, A. C.: 1982, ‘The satisfaction and happiness of some senior citizens in rural Ontario’, Social Indicators Research 11, pp. 1–30. Michalos, A. C.: 1983, ‘Satisfaction and happiness in a rural northern resource community’, Social Indicators Research 13, pp. 224–252. Michalos, A. C.: 1985, ‘Multiple discrepancies theory (MDT)’, Social Indicators Research 16, pp. 347–413. Michalos, A. C.: 1988, ‘A case for a progressive annual net wealth tax’, Public Affairs Quarterly, 2, pp. 105–140. Reprinted in Michalos 1995. Michalos, A. C.: 1991, Global Report on Student Well-Being, Volume 1: Life Satisfaction and Happiness (Springer-Verlag, New York). Michalos, A. C.: 1991a, Global Report on Student Well-Being, Volume 2: Family, Friends, Living Partner and Self-Esteem (Springer-Verlag, New York). Michalos, A. C.: 1993, Global Report on Student Well-Being, Volume 3: Employment, Finances, Housing and Transportation (Springer-Verlag, New York).

PUBLIC SERVICES AND THE QUALITY OF LIFE

153

Michalos, A.C.: 1993a, Global Report on Student Well-Being, Volume 4: Religion, Education, Recreation and Health (Springer-Verlag, New York). Michalos, A. C.: 1994, ‘Goal formation, achievement and their consequences for residential satisfaction and mobility’, in J. Cecora (ed.), Changing Values and Attitudes in Family Households with Rural Peer Groups, Social Networks, and Action Spaces (Society for Agricultural Policy Research and Rural Sociology, Bonn) pp. 31–40. Michalos, A. C.: 1995, A Pragmatic Approach to Business Ethics (Sage Publications, Thousand Oaks). Michalos, A. C.: 1996, ‘Aspects of the quality of life in Prince George: A case study in Canada’, South Asian Journal of Psychology 1, pp. 45–70. Michalos, A. C.: 1997, Good Taxes: The Case for Taxing Foreign Currency Exchange and Other Financial Transactions (Science for Peace and Dundurn Press, Toronto).

College of Arts, Social and Health Sciences University of Northern British Columbia Prince George, B.C. e-mail [email protected]

154

ALEX C. MICHALOS AND BRUNO D. ZUMBO

APPENDIX 1 Satisfaction and Happiness Levels in Prince George 1994 and 1997∗ Satisfaction with Libraries Parks Playgrounds Social/recreation facilities for seniors Recreation activities/youth Recreation activities/adults Sports facilities Support/organized athletics Garbage collection Water supply Sewage treatment Facilities for music, theatre and art Support/public performances/music/theatre Police protection Fire protection Road maintenance & repair Snow control/removal Street lighting Storm drainage Sidewalks Public transit Land use planning Enforcement/land use bylaws Animal control Landscaping blvds & other public lands Promotion of the city Your house, apartment, mobile home Your neighbourhood as a place to live Prince George as a place to live Your family relations Your living partner Your job Your friendships Your health Your religion or spiritual fulfilment Your overall standard of living

1997

1994

5.4 5.3 4.9 4.2 4.2 4.5 5.0 4.8 5.5 5.5 5.4 4.9 4.6 4.8 5.4 2.7 4.4 4.4 4.5 3.8 3.9 4.0 3.9 3.9 4.4 4.3 5.4 5.5 5.4 5.9 6.1 5.4 5.9 5.6 5.4 5.5

5.3 4.9 4.6 na na na na na 5.1 na 4.9 na na 4.4 na 3.6 na 4.1 na 4.2 3.9 4.0 na na na na 5.8 5.7 na 5.9 6.1 5.4 5.8 5.6 5.4 5.5

155

PUBLIC SERVICES AND THE QUALITY OF LIFE

Satisfaction with Your financial security Physical beauty of Prince George Your recreation activities Your self-esteem How you feel about life as a whole Hospitals and clinics here Friendliness of neighbours Ambulance service Federal government officials Provincial government officials Local government officials Shopping facilities here Elderly care services Child care services Your overall quality of life Job opportunities Traffic congestion Noise levels here Drug and alcohol abuse Number of motor vehicle accidents Weather most of the time Prince George’s geographic location Overall happiness Your accomplishments City of Prince George Elementary education College of New Caledonia University of Northern B.C. Public health service Local housing conditions Public housing Quality of water Amount of taxes you pay Judicial services Correctional programs ∗

N = 715 in 1997, 501 in 1994 na = question not asked that year

1997

1994

4.8 4.4 4.9 5.6 5.8 4.1 5.2 4.9 3.5 3.4 4.1 4.4 4.0 4.3 5.6 4.0 4.3 4.3 2.7 2.9 4.0 4.6 5.6 na na na na na na na na na na na na

4.7 4.2 5.0 5.5 5.6 3.6 5.2 4.5 3.0 3.0 3.5 4.0 3.8 4.0 5.7 4.1 4.6 4.2 2.5 2.9 4.4 4.7 5.6 5.2 5.5 4.0 4.6 4.7 4.3 4.4 3.9 4.6 2.9 3.2 3.8

156

ALEX C. MICHALOS AND BRUNO D. ZUMBO

APPENDIX 2 Domains in which there was a statistically significant difference in the levels of reported satisfaction between the 19–39 age group and the 40 and above age group at the 95% level of confidence or better.∗ Satisfaction with: Seniors’ recreational facilities Youth recreational activities Adult recreational activities Sports facilities Support for organized athletics Garbage collection Fire protection Road maintenance and repair Snow control and removal Storm drainage Land use planning Enforcement of land use bylaws Animal control Landscaping of blvds/public lands Promotion of the city Your house, apartment, mob.home Your neighbourhood as living place Prince George as a living place Your job Your financial security Physical beauty of Prince George Hospitals and clinics here Friendliness of neighbours Ambulance service Federal government officials Provincial government officials Local government officials Shopping facilities here Elderly care services Number of motor vehicle accidents Weather most of the time ∗

19–39 years

40 years and older

4.1 4.0 4.3 4.7 4.6 5.3 5.3 2.5 4.1 4.4 3.8 3.8 3.8 4.2 4.0 5.2 5.3 5.2 5.3 4.6 4.1 3.9 5.0 4.6 3.2 3.1 3.8 3.9 3.8 3.1 3.8

4.3 4.3 4.7 5.2 4.9 5.7 5.5 2.8 4.6 4.7 4.1 4.0 4.1 4.6 4.5 5.6 5.7 5.5 5.6 4.9 4.6 4.2 5.3 5.0 3.6 3.5 4.3 4.6 4.2 2.8 4.1

N = 270 for the 19-39 age group, and N = 411 for the 40 and above age group.