6/20/18

How many cars have those long old school antennas (OSA)? You know, those long antennas, not the little knubs or "shark fin" types of modern antennas. Note that by "car" I really mean "automobile". Here is an example of the type of antenna I am referring to

I wrote the following interactive program for my calculator that keeps track of a (the event, in this case, if I see a car with an old school antenna), n (if any car is observed), and a/n (the number of cars with old school antennas divided by the number of cars), and does this by me just pressing two buttons, an ENTER key and a "(-)" key:

transect()

Prgm

ClrIO

{}->ta

0->a

0->n

Lbl a

0->k

While k=0

getKey()->k

EndWhile

If k=13:n+1->n

If k=13:0->ta[n]

If k=173:a+1->a

If k=173:n+1->n

If k=173:1->ta[n]

ClrIO

Disp "n: "&string(n)

Disp "a: "&string(a)

Disp "a/n: ",approx(a/n)

Goto a

EndPrgm

This way I could just carry the calculator casually around at my side and press the keys without anyone knowing what I was doing. I also stored the vector of a's that I called "ta". I could then calculate a/n = sum(ta)/dim(ta) = mean(ta) at any time, not just see it on the screen at the time of the trial. I called this program "transect" because I plan to use it, or a modification of it, for various fun transect sampling work I am thinking about doing.

A total of 4 trials were carried out, each on a different level of a parking structure over two days. I figured that this way I'd be sure to get a reasonably fair mix of cars. I also measured my distance walked using the distance measurement tool on Google Maps. Note that each of the 1,117ft represents one lap around a floor of a parking structure. The work overall (programming plus walking) took less than 1hr.

Here are the results from the 4 trials:

- 6/18/18, Distance (ft): 1,117, OSA: 5, Total Cars: 65, OSA/Cars = ~7.7%, OSA/Distance = ~<.01/ft
- 6/18/18, Distance (ft): 1,117, OSA: 11, Total Cars: 130, OSA/Cars = ~8.5%, OSA/Distance = ~.01/ft
- 6/19/18, Distance (ft): 1,117, OSA: 12, Total Cars: 178, OSA/Cars = ~6.7%, OSA/Distance = ~.01/ft
- 6/19/18, Distance (ft): 1,117, OSA: 12, Total Cars: 171, OSA/Cars = ~7.0%, OSA/Distance = ~.01/ft

Here are some questions/comments I have. First, I expect the percent of cars that have an old school antenna to decrease to ~0% over time. The 7% number was actually much higher than I thought it would be. Second, I'd like to do trials, and many more of them, in really big parking lots, like in packed mega-store parking lots for example. Third, an old self-defense chestnut is that 'oh you can just rip off a car antenna to defend yourself'. Despite it probably taking a long time to bend the metal back and forth until it breaks (and seconds are infinity during a self defense situation), I showed even the frequency of such an encounter is too small to rely on. Last, as I mentioned already, I plan to use a similar program for investigating other questions I have that can be explored by using very simple relative frequency techniques.

Thanks for reading.

If you enjoyed *any* of my content, please consider supporting it in a variety of ways:

- Check out a random article at http://statisticool.com/random.htm
- Buy what you need on Amazon from my affiliate link
- Share my Shutterstock photo gallery
- Sign up to be a Shutterstock contributor