The Department of Transports Investment Strategy

Background

The Cycling and Walking Investment strategy from the Department of Transport clearly mentions, “We want to make Cycling and walking the natural choices for shorter journeys, or as a part of a longer journey. As part of our aim to build a society that works for all, we want more people to have access to safe and attractive routes for cycling and walking by 2040. For those studying this topic, seeking UK dissertation help can provide valuable insights into how these initiatives impact communities and urban planning.

However, according to the Department for Transport, Great Britain (2016), the pedestrian casualties in Britain had fallen in 2015 as compared to 2014 to 408 people. Despite the fall, this was still higher as compared to the number recorded in 2013, which was 393. Therefore, the pedestrian casualties seem to be still rising as compared to previous years. Despite the fact that the increase, in percentages, is statistically insignificant, it is an indication that the control and management of road networks to reduce pedestrian fatalities is not efficient. The history above as regards to the fluctuating number of pedestrian casualties implies that the risk of a pedestrian to be a casualty when a road accident occurs is increasing. Furthermore, according to Worley (2016), road traffic accidents are the leading cause of death by injury worldwide. The road traffic accidents have also been ranked the tenth among the leading causes of all deaths globally (Worley, 2016). Therefore, this is a part of the transport sector that in one way or the other, can be considered a significant portion of the global burden of ill-health (Worley, 2016). He further predicts that, if the current rates of accidents globally continue as they are, the injuries resulting from road accidents will be the third among the leading causes of global burden of injury and disease.

Studies show that Men do three times as many cycling trips as women and travel more than four times (cyclinguk.org, 2017). Similarly, the percentage of woman holding the driving license is on rise when compared to men (Dailymail, 2012). While the Government wants to promote Walking and Cycling, it is important to understand which gender is at greater risk than the other, to be a casualty when a road accident occurs and whether they were on motor or when walking & cycling. So, this study considers only two modes – Motorists and Non Motorists (Cyclist and Pedestrians) and its significance on Sex of Causality (Male and Female).

Therefore, the Null Hypothesis is “Both Men and Women are equally prone to causalities irrespective of whether they are Motorists or Non-Motorists (Pedestrians and Cyclists)”.

Data extraction/collection

The study considered data from STATS19 (on casualties of road accidents recorded within the period 2005 to 2014) where the most important variables were the sex of the casualty, casualty type, and casualty reference. Both the T-test and Chi-Square tests were considered for data analysis.

Procedure for data analysis

Using SPSS, the independent samples T-test was conducted. The focus was to test the hypotheses explained in the section that follows. As will be discussed later, the T-test assumes that the population variances between the two groups under consideration are equal (Cramer, 2003). Despite making the assumption, where this was not met, a special kind of T-test was used; where the variances under consideration were not assumed to be equal (Field, 2013). As will be identified at a later stage in this report, this was not problematic to the results of the test because no much difference was identified.

It is also important to note that the T-test under consideration has some limitations as regards it uses. According to Sheskin (2000), the independent samples T-test is limited to only comparing two samples and no more. Despite the limitation, it should be noted that this does not affect the current study because only the means of two samples are being compared.

Hypothesis testing

The Statistical testing has been carried out in two strategies

The T-test was concerned with the question; are men at more risk of being casualties in road accidents as compared to women?

The Chi-Square tests also provided the opportunity to make the research more interesting and lead to the null hypothesis; “Both Men and Women are equally prone to causalities irrespective of whether they are Motorists or Non-Motorists (Pedestrians and Cyclists)”.

T-Test for Male and Female:

This involved determining if the population means for two groups (men and women) were statistically different at the 0.05 significance level (alpha=0.05) or 95% confidence level. For the T-test, the hypotheses were set as follows;

H0: µWomen - µMen = 0 (“The difference of the means is equal to zero”)

H1: µWomen - µMen ≠ 0 (“The difference of the means is not equal to zero”)

The µWomen and µMen represent the population means for women and men respectively.

Before testing the hypothesis above, it is important to consider the assumption of homogeneity of variance at the 95% level of confidence (alpha=0.05). This implies that both men and women have the same variance. This was appropriate because the independent samples T-test, as stated before, would always require the assumption that there is the homogeneity of variance between the two groups under consideration (Sheskin, 2000). The said assumption was tested by way of SPSS including the test for homogeneity of variance, the Levene’s Test while running the said independent samples T-test. Therefore, there had to be considered the hypotheses for Levene’s test which were as follows.

H0: σWomen2 - σMen2 = 0 ("The population variances of women and men are equal")

H1: σWomen2 - σMen2 ≠ 0 ("The population variances of women and men not equal")

With σWomen2 and σMen2 representing the population variance of women and men respectively.

The “Group Statistics” above also provide the descriptive statistics for each of the two groups under consideration (Men and Women). These are defined by the grouping variable (sex_of_Casualty). The table indicates that there are more men (107556) than women (73776) captured by the data.

The inferential statistics are provided by the "Independent-Samples T Test" table above. These provide a variety of tests including the "Levene’s Test for Equality of Variances” which provide information about the assumptions of the t-test. As discussed before, the assumption by the t-test is that there is approximately equal variability between the two groups, women and men. The Levene’s Test’s significance results provide a p-value = .001. Thus, there was a significant difference in Casualty_Reference between men [M=1.35, s=1.057] and women [M=1.51, s=1.175] Based on the results, the sig-value = .001 which is less than the set significance of level (0.05).

Chi Square Test – Null Hypothesis:

The casualty type was grouped into Motorised and Non-motorised. The Non-Motorised includes Pedestrians and Cyclists. Motorised include all other categories.

The Null Hypothesis is set at

µ0 = Men and Women are equally prone to causality in both Motorised and Non-Motorised mode.

µ1 = Rejection of Null Hypothesis.

The chi-square results are as below.

Based on the table above, sig 0.001<0.05, it can be concluded that there is a statistically significant relationship between gender as regards to Motorised and Non-Motorised modes.

Discussion

Based on the results of the two tests, the null hypothesis should be rejected. Calculating the mean difference, for the T-test, it was found out to be -0.16 (1.35 – 1.51), an indication that women feature more within a particular Casualty_Reference. The results above give the implication that women are more at risk of being casualties in a road accident as compared to men. Furthermore, the Chi-Square test results indicate that the use of the different modes of transport is statistically significant based on gender.

From the interpretation of the T-test results above, where the mean represents a particular reference to a casualty. It is important also to note that the hypotheses being tested presented the opportunity for understanding that there exist statistical significance between men and women involved in road accidents as casualties.

According to Brake-UK (2016), statistics on road safety indicate that there is a big difference between men and women who are involved in car accidents as pedestrians. A close look into the Chi-Square test reveals that Men are more prone to causality while walking and cycling. Similarly, Women are more prone to causality when they are travelling in a Motorised vehicle.

The WHO (2004) identifies some risk factors which are attributable to high rates of pedestrian casualties on road accidents. The risk factors influencing crash involvement include the presence of fatigue caused by alcohol, medicinal, or other drugs, excessive or inappropriate speed, being a young male, and being a vulnerable user of residential or urban roads. Other risk factors are poor road user eyesight, travelling in darkness, and other vehicle factors including braking, maintenance, and handling. These are factors, which influence both women and men and thus cannot be fully depended upon to determine who, between men and women, are at a greater risk of being casualties of road accidents. When these factors are studied together with what Brake-UK (2016) explains, that women make fewer Cycling trips compared to Male, it will be learnt that men are exposed to more casualty risk than women. Further statistics from DfT-Great Britain (2015), Brake-UK (2016a), and DfT-Great Britain (2016) provide consistent results that men are more likely to be at a higher risk of being casualties, at even severe stages (more serious casualties) than women.

Conclusion

The discussion above indicates that the hypotheses test results of the Chi-Square are consistent with the existing literature on road safety in Britain. Also, future hypotheses tests for cases such as this one should consider other tests including F-tests, regression modeling etc.

Bibliography

Brake-UK, 2016a. UK road casualties. [Online]

[Accessed 13 December 2017].

Cramer, D., 2003. Advanced Quantitative Data Analysis. England: Open University Press.

Daily Mail. 2017. Daily Mail. [ONLINE] Available at:

Department for Transport, Great Britain, 2016. Reported Road Casualties Great Britain (RRCGB), London: Crown.

DfT-Great Britain, 2015. Facts on Road Fatalities, Landon: National Statistics.

DfT-Great Britain, 2016. National Travel Survey: England 2015, London: National Statistics.

Field, A., 2013. Discovering Statistics using IBM SPSS Statistics. 4th ed. s.l.: Sage Publications Ltd.

Sheskin, D. J., 2000. Handbook of Parametric and Nonparametric Statistical Procedures. 2nd ed. Boca Raton: Chapman & Hall/CRC.

Continue your journey with our comprehensive guide to Statistics For Travel And Tourism.