Chapter 2. Research Method

This report is based on data collected from Montana, North Dakota, and South Dakota rural road users and providers' (also referred to as decision makers) perceptions of township and county roads. The perceptions were attained through a questionnaire mailed to various user groups and decision makers in the year 2000. It could be assumed that users and decision makers agree on the condition of the roads if their perceptions closely match. Furthermore, if providers know what the road users need, it would be easier to make better user-based decisions. In addition, it would be beneficial for decision makers to share relevant information with the road users and ask for their input because this would create a better partnership or alliance between them. The method used to attain road user and decision makers' perceptions about the road system are explained in this chapter.

Data Collection

To better understand the perceptions of road users and decision makers, primary data were collected by a mail survey. The survey was sent to select road users and decision makers to gather pertinent attitudinal information. The survey and the methods used in this comparison will be examined more closely in the following pages.

User Group Identification

The LTAP Centers and DOTS in each state helped identify the specific user groups to survey. The specific user groups vary among the states. In Montana, the user groups surveyed are school bus and transit drivers. The North Dakota user groups are agricultural producers, school bus drivers, and rural road commuters. The South Dakota user groups are delivery services and mail carriers.

Survey Instrument Design

A two-page survey was developed and mailed to selected user groups and decision makers to compare their perceptions. The questionnaire contained only 12 questions to assure as many responses as possible. Questions were divided into sections relating to physical roadway conditions, operational conditions, maintenance, funding, and needed improvements and were kept uniform among the different questionnaires. Several questions asked for a YES or NO response along with a brief explanation, while other questions had a five-point scale used for rating each roadway factor along with a "not applicable" rating. The rating range was 1 = very good, 2 = good, 3 = fair, 4 = poor, 5 = very poor, and 6 = not applicable. The final question on the survey asked road users and decision makers to list the 10 improvements they would like to see on roads they most frequently travel.

A county road advisory committee pre-tested the survey for relevance of issues and ease of completion. The survey instruments were modified to incorporate the suggested improvements.

Mailings

The LTAP Centers and DOTs worked to obtain the mailing lists and labels for the user groups from private and public sources. The response rates for each of the states are presented in Table 1. In Montana 500 surveys were mailed with 238 returned for a 48 percent response rate. In North Dakota 1,927 surveys were mailed with 473 returned for a response rate of 25 percent. South Dakota sent 688 surveys with 377 returned for a 55 percent return rate. The overall return rate for the tri-state area was 35 percent.

Table 1. Response Rate of Groups Surveyed in the Tri-State Area

Responses from the survey were entered into a spreadsheet. The rating structure was combined for road services and features for the local road system. Before any analysis was performed on responses for road services and features, certain ratings were combined. The ratings were between 1 and 5 (1=very good; 2= good; 3= fair; 4=poor; 5=very poor 6=not applicable). For analysis purposes, ratings 1 and 2 were combined to represent "good" while 4 and 5 were combined to represent "poor." A further expansion of the survey analysis included a breakdown of suggested road improvements. Each road user and decision maker listed up to 10 road improvements they would like to see on the roads they most frequently travel. The rural road users were combined in each state to make the comparisons to decision makers more straightforward. Appendix A contains tables with the results of each user group by state. The research methods used to analyze the data were straightforward. A chi-square test was used to identify statistical significant differences in perceptions between the rural road decision makers and rural road users. This test would indicate if the decision makers and the users view roadway elements and services similarly. In addition, basic means and frequencies were calculated for the variables to compare mean values and percentage responses. A significant difference was found between several of the road and service features. These differences are identified in the next chapter.