**2. Materials and Methods**

The main aim of this research was to assess the behaviour of pedestrians and drivers and the pedestrian–driver relationship at and in the area of pedestrian crossings. Such systematic studies are necessary due to the level of risk of pedestrian accidents. The objective was to make walking more attractive and improve pedestrian safety. To this end, field tests and survey studies were carried out in 70 test points in Poland's Pomorskie and Małopolskie regions, across an area that is diversified in terms of its development. The road cross-section and location of pedestrian crossings were also taken into consideration. The results of these studies make it possible to assess pedestrian safety at various types of crossings, associated with the street's cross-section.

#### *2.1. Measurement Technique*

The following instruments were used to record traffic parameters alongside pedestrian and driver behaviour:


Depending on the test site, one or two video cameras were used to record images. The cameras were situated on a tripod at a distance of 10 m to 15 m behind a pedestrian crossing and located on the shoulder, lamppost or pavement along the road. The measuring equipment was located in a way which would not capture the attention of drivers or pedestrians. Driver behaviour was studied as cars approached the pedestrian crossing. The maximum range of image recording depended on road geometry and was between 40 m and 100 m. During the measurement, test cross-sections were designated in the field, every 10 m or 20 m, as a test basis for the assessment of average speed (Figure 2). The speed was calculated on the basis of the length of the measured segment and time travelled to

the test cross-sections. Image analysis covered recording the time of consecutive events, i.e., the appearance of vehicles in the subsequent test cross-sections. Speed measurement included only the vehicles moving in free traffic, i.e., those which were not influenced by other vehicles in their vicinity. The following scenarios were selected in the analysis of vehicle speed on a pedestrian crossing:


**Figure 2.** Diagram of field tests using video recording instruments.

For pedestrian traffic parameters, the camera's field of observation covered the entire pedestrian crossing under analysis and its access path. The analysis included the following:


Each test included pedestrians going across the crossing in both directions, with a division into age groups also included (<20 years—11.1%; 20–60 years—80.4%; >60 years—8.5%). In the case of drivers, age was not taken into account.

#### *2.2. Selection of Test Sites*

When selecting the test points, limitations were taken into account, making it possible to obtain a random sample of test points with a high uniformity of traffic conditions. This allowed a wide range of comparisons between test results. At each pedestrian crossing, the measurements were carried out for 6 h (10:00 a.m.–3:00 p.m.), in good weather conditions and in daylight. The research was conducted only on working days (Monday–Friday), in the months of April–June. In order to obtain such a set, test points were located as follows:


In order to compare vehicle speeds and pedestrian behaviour, selection criteria were designated for the test points. The following criteria were chosen:

	- Field tests with pedestrians were carried out in three types of areas:
	- In each of these areas, tests were performed in the following locations:

For the locations selected above, detailed locations were specified, taking into account various cross-sections of the road segment:


The current speed limit was identified for each of these test point's selection criteria and, for further analysis, 50 km/h (for built-up areas) and 70 km/h (for built-up areas with dual carriageways and for rural areas) were selected. Figure 3 presents a cumulative diagram of the test site.

**Figure 3.** Diagram of test point selection.

#### *2.3. Statistical Methods*

Statistical methods were reduced to calculation descriptive statistics according to pedestrian and vehicle speed. Average speed, standard deviation and the max and min speed were calculated. Average speed and differences in speed for various locations of pedestrian crossings and the presence of pedestrians could be factors to assess the interaction between drivers and pedestrians. Therefore, in Section 3, results of the average speed comparison are presented. Additionally, to assess the statistical significance of speeds, a two-sample *t*-test for the difference of means was carried out. To assess speed distribution, the Shapiro–Wilk test was used to assess normal distribution.
