Section 1: Fatal Accidents

Introduction

We’re going to examine different categories of vessel operators found within BARD that correspond to fatal accidents. Primarily focusing on boating safety education, let’s begin by getting an idea of the operator education makeup for the 645 recreational boating deaths that occured in 2019.

## 
##           Informal           Internet       No education        No operator 
##                 15                 11                208                 11 
##              Other              State            Unknown US Power Squadrons 
##                  6                 54                331                  1 
##     USCG Auxiliary 
##                  8

Given what we know about sample size, we shouldn’t even attempt analyzing any category with less than 30 observations. Therein, today we’ll focus on:

  1. No Education
  2. State Education

Unfortunately, the largest category is Unknown. While it has a robust sample-size, we can’t examine it for any takeaways, as we have no idea what type of education (if any) are found within that sub-group.

However, I’m content to press on with a comparison of No education to State education. If we find that state-based education has different types of accidents occuring proportionately when compared to no education–we may be able to improve something relatedly moving forward.

Key Takeaways

1. Cause of Death

  1. Those with no education had 7.9 percent more drownings listed as the cause of death within their group, compared to those that took a state-based boater education.

  2. Those with state-based education had 10 percent more ‘trauma’ listings and five percent more listed as ‘hypothermia’ proportionately, compared to those with no boater education.

2. Water Conditions

Those with state-based education were proportionately more likely to have fatal accidents in challenging water conditions compared to those with no boating safety education. Perhaps, those with state-based education felt more capable relative to the water conditions than those with no education; lending to a (potentially) false sense of security while on water.

3. Primary Accident Event During Fatal Incidents

  1. The largest finding is that those with no education had capsizing events make up 20 percent of their group’s total fatal accidents. Operators with state-based education capsized in fatal accidents in only 3.7 percent of their cases.

  2. Those with state-based boater education were three percent less likely to fall overboard proportionately and 4.5 percent less likely to have flooding/swamping, than those with no education.

  3. Those with no boating safety education were 10.4 percent lower in terms of the share that collided with another recreational vessel, 7.6 percent lower for having had a person depart from the vessel, and 3.5 percent lower for having a person ejected from the vessel.

  4. Our group with no boating safety education had a 10.4 percent lower share of individuals collide with another recreational vessel.

  5. Within this same group, 7.6 percent and 3.5 percent less had a person depart from the vessel or be ejected from the vessel respectively.

4. Primary Accident Cause During Fatal Incidents

  1. Operators with state-based education had a 10.8 percent higher share of cases with alcohol listed as a primary fatal accident cause, compared to those with no boating safety education.

  2. Similarly, operators with state-based education had a 7.7 percent higher share listing operator inattention, 4.9 percent more listing hazardous waters, and three percent more listing excessive speed listed as the primary fatal accident cause.

  3. Operators with no experience had 4.4 percent more cases within their group with operator inexperience listed as a primary accident cause, compared to those with state-based education.

Section 2: Injuries

Introduction

Now, we’re going to look at accidents where at least one injury occured. In doing so, we’ll replicate the same tables and charts as rendered above for fatal accidents. Let’s begin by getting an idea of the operator education makeup for the 2,696 recreational boating injuries that occured in 2019.

## 
##           Informal           Internet       No education        No operator 
##                112                 99               1056                 63 
##              Other              State            Unknown US Power Squadrons 
##                 57                476                750                 14 
##     USCG Auxiliary 
##                 69

Let’s focus on the categories with the most observations.

  1. No Education
  2. State Education

Key Takeaways

1. Type of Injury

The types of injuries that occurred were pretty similar proportionately between our no boating safety education and our state-based groups. The only noteworthy distinction is that operators with a state-based education had a 8.8 percent higher share of cases with a cut injury.

2. Water Conditions

Unlike what we saw in Section 1 with fatal accidents, water conditions when an injury occured were very similar between our groups.

This undermines our earlier hypothesis that, perhaps, those with state-based education felt more confident in rougher waters due in part to a ‘false sense of security while on water’.

3. Primary Accident Event During Injury Incidents

Again, very little discernible differences between our groups. The types of accidents as they occur proportionately are nearly identical.

4. Primary Accident Cause During Injury Incidents

While almost all the accident causes were remarkably similar proportionately between our two groups with regard to injuries, one thing stands out.

Those that received state-based boating safety education had 4.8 percent more individuals with improper lookout listed as the primary accident cause, compared to those with no education.

Section 3: Conclusions

  1. Its important to note that we are not talking about probability or predictions. When we say, “Group 1 had x percent more than Group 2…”, we are speaking in terms of the proportional makeup of their group as when compared to another. To make predictions for the future (ex. If one has no boating safety education, what’s the likelihood that their boat capsizes compared to someone who took a state-based course), regression analysis must be pursued. To do this properly, we would need to build a model in the future with many more variables (for the sake of trying to capture every possible factor that might affect outcomes).

  2. That being said, in terms of descriptive statistics, the most noteworthy differences between our groups are captured in Section 1 (Fatal Accidents) above.

  3. Section 2 (Injuries) found very similar proportions between our groups in almost all of the figures presented.