Looking at Idaho

We looked at Pennsylvania in terms of common reasons people believe the death toll has gone up. Now we move on to Idaho.

Idaho is a small state—it’s entire population is just over 1.5 million. It also only began its state training program in 1996. Small changes in registration, training and fatality numbers, then, can make huge differences in percentages. In these ways, it’s not a good comparison to Pennsylvania.

Otoh, it provides a good comparison to Pennsylvania because it taught the MRC:RSS in the classroom until 2006 and then used TEAM Oregon’s BRT, and the RSS range until 2007 when it switched to the BRT range. This then does not give us enough passage of time to see if there was any change with the BRT. Otoh, it gives us one point of comparison on the effect of a curriculum and/or the state program administration:

Motorcycle registration

In 1997, motorcycle registrations were 34,935 and in 2007 they had increased by 9,012 to 43,947 or a 25.8% increase. In four out of the 11 years, however registrations decreased and rebounded. However, two of those decreases were in the past two years—the high for Idaho motorcycle registration, then, was in 2005 with 56,402.

Helmets

Idaho does not require a helmet for those 18 and older—this was the law before the training program began and is still the law.

Training

The state program began in 1996 so we do know exactly how many students were trained prior to 1997—264—although we don’t know how many riders were trained in other states.

Training went up 661.7% from 1996 to 2007. If 2008 is included, it went up 815.5%. But remember the curse of small numbers resulting in large percentages. At the end of the 2007, a total of 14,456 had taken basic training. If we assume all students went out and bought a bike and rode it, then just over 38% of Idaho’s riders were trained by the end of 2007.

Idaho is one of the very few states that actually found out how students performed on the road after graduating and so we don’t need to try to figure it out in other ways. One caveat—the numbers include those who took the basic and experienced rider courses and its impossible to differentiate:

Fatalities

From 1997-2007, fatalities went up 61.11%, which is to say it rose from 18 to 29 with a grand total of 204. However, in four of those years, fatalities were equal to or less than 18.

Before we get into the specifics, let’s look at Idaho’s training picture:

ear

Fatalities

Total Basic Training Students

Passed

Failed

Dropped

Did Not Graduate

1997

18

620

84.80%

8.20%

6.90%

15.10%

1998

6

544

88.70%

5.69%

5.14%

10.83%

1999

13

839

85.22%

9.17%

5.60%

14.77%

2000

18

1003

87.53%

5.78%

6.67%

12.45%

2001

19

1129

85.29%

8.59%

6.11%

14.70%

2002

12

1265

84.55%

7.74%

7.58%

15.32%

2003

19

1431

84.06%

8.31%

7.54%

15.85%

2004

24

1681

87.44%

7.37%

5.17%

12.54%

2005

26

1929

85.58%

7.87%

6.48%

14.35%

2006

38

1740

81.95%

10.63%

7.41%

18.04%

2007

29

2011

87.41%

8.50%

4.07%

12.58%

In Pennsylvania, the percentage of students that passed rose when the BRC began statewide in 2002 but the dramatic change was in the fail rate, which dropped to single digits. In contrast, Idaho’s percentages are consistent in all four areas (passed, failed, dropped and the combined “did not graduate”). Like Pennsylvania, the fail rate was in the single digits but unlike that state, in Idaho it was so in all but one year. However, Idaho’s averaged fail rate is 7.99% and Pennsylvania’s averaged BRC fail rate was 5.58%, and Idaho’s combined “did not graduate” and is slightly higher than Pennsylvania’s as well.

In Pennsylvania, there were two fatalities and one resulting in complete paraplegia from crashes in training, during this time. And, according to the state coordinator, there were incidents when bystanders were hit as well as students hitting sheds, cars and motorcycles off-range.

In Idaho, there have been no such incidents.

It is unknown if trained motorcyclists ride more or less or the same amount as untrained riders—or how many trained motorcyclists went on to ride at all after passing the course—or how many of those that did not graduate went on to ride.

Fatalities

From 1996-2007, only 22 students (10.78%) trained in Idaho were killed in a motorcycle crash.

When it comes to kind of crashes, 68% of trained students were killed in multi-vehicle crashes (MVC) and 32% in single-vehicle crashes. In comparison, 43% of untrained riders died in MVCs and 57% died in SVCs.

Impaired (alcohol/drugs) fatalities were 31.37% of the total. Trained rider fatalities comprised 4.69% of that total and  95.31% were untrained.

Just over 54% of trained rider fatalities were wearing helmets while almost 81% of untrained rider fatalities were helmeted.

When it comes to injury crashes, trained riders were also about the same percentages.

In 2004, the Idaho STAR program hired Bill Bohley, past NHTSA Associate Administrator for Research and Development to analyze the crashes from 1996-2003 and it was redone in 2007. The analysis included ERC students and assumed that all students owned a motorcycle and all riders—trained and untrained—rode the same amount. Based on those rather dubious assumptions and some simple calculations, the analysis concluded, “…training is associated with of reduced crash risk of 71%…and… an 81.4% reduction in the risk of a fatal crash”.

Because we know more than we did in Pennsylvania, there’s no need for the training vs. fatalities graphs.

Motorcycle Registrations v. fatalities

When it comes to motorcycle registrations, we find the same thing as we did in Pennsylvania—there appears to be little or no correspondence:

ID Motorcycle Registrations v. Fatalities

On the surface it appears there’s even less correspondence between training and motorcycle registrations in Idaho:

ID MC Registrations v. Aggregate Training

But it’s worth looking a little deeper: If we assume that each motorcycle represents one owner and all students end up buying motorcycles as Bohley did, then in 1996, 33,646 motorcycles were owned by untrained (or not trained in Idaho) riders.

If we compare the untrained riders in 1996 to the number of untrained riders in 2007 then that number falls to 29,491 (-12.3%). We don’t know if that represents people who stopped riding for one reason or another or if some of them took training.

In 1996, only 0.77% of motorcycle registrations represented trained riders,  but by 2007, the aggregate of trained riders had increased to just under 33%. But if we compare the aggregate of trained riders to the original number of registered motorcycles, it’s just over 52%.

The hidden number, though, is that only 4,155 (9.45%) of the 2007 motorcycle registrations are not accounted for by the (estimated) original and untrained owners and the aggregate of trained students. It could be, then, that only 19% of Idahoan motorcyclists registered a motorcycle and had not taken training at some point.

We don’t know if Bohley was correct in his assumption—other research, including Billheimer, suggests that 5% of trained students choose not to ride after the course and research suggests that trained new riders don’t ride as much as more experienced riders. Additionally, it’s reasonable to assume that many owners bought an additional motorcycle during those years.

While leaving mileage—and therefore exposure out of it—we can adjust for Bohley’s other assumptions: If Billheimer’s 5% do not ride and MIC’s owner to motorcycle formula is applied, trained riders represent 57.4% of all owners.

This strengthens the suggestion that training has become a normative part of buying and owning a motorcycle in the past 11 years.

Depending when those 22 fatalities occurred—and that information I don’t have—the success of the Idaho STAR program may be greater than Bohley’s analysis suggested. Otoh, it might not.  More study is needed.

Helmet use

Idaho doesn’t require helmet use for riders 18 and older. The number of riders killed wearing a helmet has been trending up in the past four years while the percentage of unhelmeted fatalities is decreasing. NOPUS surveys have found helmet use increasing in western states and the Idaho DOT reported that in 2005, 64% of riders involved in collisions were wearing helmets. However, Idaho STAR’s analysis states only 38.2% of all collision-involved riders were wearing helmets.

ID Helmet use v. fatalities

Conclusion

Idaho STAR is to be praised for collecting the kind of data every program should. That effort was initiated and the first analysis was done under Ron Shepherd’s leadership and has been continued under his successor, Stacey (Ax) Axemaker.

It’s not enough to draw any conclusions about the efficacy of training—there’s too many assumptions that have to be made and it also depends on other unknowns. For example, it comes back to the self-selecting process: do those who tend to greater safety efforts and caution choose training and helmets? Or does training change minds so they ride safer and wear helmets?

And it should be pointed out that, in lieu of needed information, the Pennsylvania data at this point shows that training went up at about the same rate as fatalities. Idaho’s statistics do show there are differences—and there are similarities. It’s unknown if any are significant. So we’ll look at some other states and see if a clearer picture emerges.

Advertisements
Explore posts in the same categories: History, Motorcycle Safety, Motorcycle Training, State Motorcycle Safety Programs

One Comment on “Looking at Idaho”

  1. gymnast Says:

    The difficulties of conducting valid research to explain the occurrence of “rare events” using small samples are better appreciated by reviewing some of the materials found by doing a Google search of “small N statistics”. Statistical research using administrative data is problematical when it comes to explaining the the behavior and behavioral motivations behind the statistics. Sometimes it is best to “let the data, in it’s simplest form speak for itself”. Idaho is a small enough state that it is relatively easy to “keep score” (though scalability should not make it any more difficult for larger states to collect like data).

    It is sad that better data is not available on the relationship between the variables of training, fatalities, voluntary helmet use, age, and time (chronological relationship) and the criterion variables of crash, injury, and injury severity index including fatalities.

    You do an amazing job of of providing a cogent analysis considering the limited resources available and at your disposal. A far better job, in fact than the organizations and institutions that claim to be the leaders and policy makers in the field of motorcycle safety.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: