Continental Research
!
4500 Colley Avenue
!
Norfolk, VA 23508
5
Executive Summary
(cont’d)
~ ~ Rating 24 City Services and Characteristics ~ ~
(cont’d)
While reviewing the percent who are satisfied with each service is interesting, a more sensitive
measure is typically used to make finer distinctions. The chart on the next page shows the
“average” satisfaction scores (i.e., the means). These means are more sensitive to the
distribution of the responses across all four levels of satisfaction.
To calculate the average satisfaction ratings, each rating was assigned a numeric value (4=Very
Satisfied, 3=Satisfied, 2=Dissatisfied, and 1=Very Dissatisfied). Using this metric, an average
rating of 4.0 would be considered a “perfect score” (i.e., all respondents responded “Very
Satisfied”). A high average rating usually reflects many people being “Very Satisfied.” (When
only the percentage who are “Satisfied” is examined, those higher marks are hidden.)
To illustrate how using the combined “Satisfied” percent can impact one’s interpretation,
consider the previous bar chart where 91.3% of the respondents were satisfied (“Very Satisfied”
+ “Satisfied” combined) with the overall level safety in their neighborhood. That satisfaction level
ranked as the 11 highest. When we look at neighborhood safety in terms of the average
th
satisfaction score (where 4=Very Satisfied, 3=Satisfied, 2=Dissatisfied, and 1=Very Dissatisfied -
see next page), it ranks higher than the other services/characteristics (avg. = 3.30 out of a
possible 4.0) because many people were “Very Satisfied.” In this analysis, the average
satisfaction score is the 3 highest rating. Again, the averages are more sensitive and better
rd
represent the full distribution of responses.
The chart on the next page displays the average (mean) ratings in descending order. The
highest possible average is 4.0 (if everyone reported being “Very Satisfied”) and the lowest is
1.0. Again, it is important to remember that some items usually rank near the top of the list
simply by their nature (e.g., fire and police services), while others are traditionally found closer
to the bottom (e.g., road and traffic issues). Presenting these services together in no way
indicates a desire to compare departments, as that would be unfair. Each provides a unique
service that should be tracked over time independent of the other departments.
Whenever averages were available to compare the survey data over time, the significance of any
differences in the averages was tested using a t-Test statistic. The t-Test is a detailed
computation that can identify when two averages (i.e., the means and their distributions) are
"significantly different" from one another. While some questions were asked in 2008 or 2007,
much of the comparative data were from 2006 when a relatively similar survey was used.
Statistically significant t-Test differences when compared to the “most recent” survey, whether
trending up or down, are indicated below.
Significant increases over most recent data available:
The Fire Department:
3.34 in 2006 3.43 in 2014
The public school system
3.01 in 2006 3.16 in 2014
The quality of the drinking water
2.70 in 2006 2.88 in 2014
The traffic flow on City roadways
2.18 in 2006 2.40 in 2014
City services (overall)
3.00 in 2008 3.20 in 2014
Significant declines from most recent data available:
The public library system
3.51 in 2006 3.40 in 2014
