An unscientific, informal modeling survey: The results are in

After seeing a few discussions recently on internet forums regarding the accuracy of a forthcoming kit, my curiosity was sparked to seek the answers to some simple questions. How many models actually get built? What is really important to modelers? The more I thought about, the greater my curiosity became. At some point, in a fit of mild insanity, I put together some survey questions, and let a few friends see it. They encouraged me with such uplifting words as “whatever ya weirdo” and “I just came to Starbucks to drink coffee, leave me alone!” So I decided to move forward. (OK- I made those statements up…. 🙂 I do appreciate the friends who were my “test subjects” and gave me real encouragement to move forward.)

I’d hoped that maybe one hundred people might respond, at best. However, I was totally amazed at seeing 575 respond to the survey. I wish to extend a great word of thanks especially to Brett Green of Hyperscale for allowing me to post a link to the survey there.

Some problems and mistakes did emerge once the survey was launched, though I didn’t feel any were so catastrophic to invalidate the process. A few questions could have been worded better. I should have included some questions for gender and world region. And I should have done a better job promoting the survey to attract modelers from other genres. (Although the efforts I did make were largely not responded to, at least that I could tell.)

And of course, the whole process is skewed online heavy because I didn’t have time to go door to door. 🙂 I’m sure there are many modelers who aren’t that engaged online, as I know a few personally. So I’d imagine there are more. But whether that represents 1%, 10% or 90% of the market, I don’t know.

So what you have here is some mildly interesting questions, graciously responded to by 575 people, and it does give us some insights into the world of modelers. The survey has its flaws and warts, and the following analysis is very amateur at best. 

If you’re curious about the results, and want to skip my admittedly amateur analysis, please- go right to the results. I’m also making the raw data available via a spreadsheet, if you’d like to examine things a bit further. Please contact me at if you’d like the data in an Excel spreadsheet. (Or Google Sheet, or CSV file.)

Please keep in mind I am no professional survey…. err… person. (I don’t even know the correct term for the field, apparently.) Anyone who is a…. survey-taker-person…. please feel free to analyze the data in a more precise way. (Or at least let me know the term for this career field.)

Good? Good.

For anyone who is left at this point, please read on. 

(Note- for each table of data, column one shows the response, column two shows the number of respondents, and column 3 shows the total percentage.)

What is your age group?

Under 18 1 0.2%
19-30 11 1.9%
31-40 30 5.2%
41-50 149 25.9%
51-60 248 43.1%
61-70 116 20.2%
71+ 20 3.5%

No surprise here. Modelers are basically middle aged to old. The largest single group in the survey was represented by the 51 to 60 age group, with the 41-50 age group coming in at at second by a fairly large margin. I suppose this should come as no surprise, as it illustrates the fact that the largest group of modelers today was also the largest group of modelers in the hobby’s heydey for youth- the late fifties through the early eighties. 

I suppose this is why today is often referred to as the “Golden Age Of Modeling”. It’s the point in time when the modelers of yesteryear have enough time and disposable income to fully pursue the hobby in ways we only dreamed of when we were 11 years old.

What is stunning is how low the numbers are for the under 40 folks. Less than 8% are in that group. And whoever the one person is under 18…. please turn out the lights when the rest of us are gone.

I guess when you get down to it, this is looking like the last decades of the hobby. Within the next 25 years (statistically speaking of course), a whopping 66% of the people in the survey will no longer be buying models, paints, decals, photoetch or participating on forums.

The model industry simply can’t go on in that situation, at least not in the form it is today. And while there are efforts to appeal to younger modelers, the replacement rate is severely negative.

It’s not to say that the companies we know and love today won’t be around. They’ll likely have to shift their focus. To what- I don’t know.

What term is closest to your *prefered* genre of modeling?

Aircraft 445 77.4%
Autos 19 3.3%
Armor 61 10.6%
Ships 23 4%
Sci-Fi 3 0.5%
Space 1 0.2%
Figures 8 1.4%
Diorama 3 0.5%
Other 12 2.1%

What genres of modeling do you typically model in?

Aircraft 523 91%
Autos 136 23.7%
Armor 293 51%
Ships 190 33%
Sci-Fi 120 20.9%
Space 59 10.3%
Figures 82 14.3%
Diorama 47 8.2%
Other 18 3.1%

I’ll admit upfront that genres outside of aircraft are grossly underrepresented in this survey. Being an aircraft modeler, my connections are generally confined within the group of people who like things with wings. The efforts I did make in outreach to other areas did not seem very successful. (Or maybe they were wildly successful and there are just a lot more aircraft modelers…)

So it’s no surprise that the numbers in Preferred Genre are so heavily skewed in towards aircraft. Still, I think the Typical Genre question does have some value in that it shows the diversity that exists even within a group that for the most part most closely identifies with a single genre. What I don’t know is if the same diversity exists among folks who primarily build cars, or tanks, or ships, etc. Certainly an interesting question to pursue in follow-on surveys.

As to the numbers, for the preferred genre, aircraft was by far the winner, with over three fourth of the votes, while armor was a distant second at around 10%. Again- these numbers are likely skewed simply because the participation was heavily favored towards aircraft. 

What is your *prefered* modeling scale?

1/144 or up 10 1.7%
1/72 104 18.1%
1/48 293 51%
1/35 52 9%
1/32 71 12.3%
1/25 9 1.6%
1/24 5 0.9%
Other 31 5.4%

What scales do you typically build in?

1/144 or up 115 20%
1/72 291 50.6%
1/48 424 73.7%
1/35 228 39.7%
1/32 250 43.5%
1/25 58 10.1%
1/24 83 14.4%
Other 85 14.8%

These two suffer from the same flaw as previous two- heavily aircraft skewed. We can pull out some interesting insights though.

For the preferred scale, 1/48 was by far the winner, with over 50% of the vote. 72nd scale fell in at 18%, and 32nd at 12%. 

When opened up to multiple answers, the breakdown again showed a great diversity in the scales modelers will build in. So while one scale was the general preference, it did not seem to be a hard “lock in” across the board.

What is the *primary* factor in your purchase of a model?

Subject 487 84.7%
Accuracy 17 3%
Manufacturer 7 1.2%
Scale 23 4%
Recommendations/Reviews 8 1.4%
Expected build experience 21 3.7%
Price 12 2.1%

The clear winner in this was subject. By a huge margin- 85%. I was not surprised that it was number one, but I was surprised by how much. I suppose my own experiences colored my expectations, as I’ve frequently decided on a model because of price, and the subject being secondary. 

And while there is no quantifiable direct tie-in to the survey, I do think this might illuminate the age old question of “do we really need another {name any type of model}?” Given that subject is by far the driving factor in purchase, and given that those Spitfires and 109s and and Mustangs sell like crazy, then I suppose the basic answer is “Yes, we do need another.” Why? Simple- they sell! (But that is another survey…)

Rank the factors in your purchase of a model

This question caused a bit of confusion with a few respondents. The idea behind it was to have each person rank seven criteria in order of importance. I realized that while some folks might see two factors as perfectly equal for them, I surmised that in order to get the best results, I needed to set the grid so only one answer was allowed at each level. Unfortunately, I didn’t do a good job of explaining it.

However, I do think that in the long run, it produced results that most accurately reflected the factors for the larger group. So if you felt two factors should have been equally weighed, please do accept my apologies. (And I do mean that…) And for those who thought each answer should be evaluated on its own merits, and not against the others, I also apologize. Despite your frustration, the method did force you to answer according to the intent of the question. 

So anyway, what do the numbers say?

No surprise that first place was Subject.

1 409 71.1%
2 81 14.1%
3 18 3.1%
4 12 2.1%
5 3 0.5%
6 18 3.1%
7 34 5.9%

As you can see it was by far the most selected category.

From there, things got a bit more diverse.

Accuracy shook out in about 2nd or 3rd, depending on how you want to average it out against all the others.

1 16 2.8%
2 156 27.1%
3 181 31.5%
4 100 17.4%
5 60 10.4%
6 42 7.3%
7 20 3.5%

This didn’t surprise me either, as I know the subject of accuracy is an oft debated one, and it’s also a very subjective one. For one modeler, accuracy may boil down to “yes, it looks like a Spitfire.” For another, the same kit may evoke convulsions. Yet both people, despite differing views on accuracy, may have selected 2nd place, or 3rd place, or whatever.

The numbers for manufacturer came out a bit different.

1 19 3.3%
2 27 4.7%
3 77 13.4%
4 125 21.7%
5 99 17.2%
6 95 16.5%
7 133 23.1%

For the most part, modelers either sorta care about manufacturer, or they don’t care at all, to make a generalization. I was somewhat surprised by this. I’d have thought the number would have been a bit more solidly centered in one region. (Of course, to be fair, the method of scoring may have been partially to blame for this. Maybe not… I don’t know.)

Scale can be fairly solidly attributed to second place overall. While accuracy had some high 2nd place numbers, scale was the clear winner.

1 70 12.2%
2 199 34.6%
3 131 22.8%
4 74 12.9%
5 47 8.2%
6 35 6.1%
7 19 3.3%

No shocker here, really, as it simply amplifies responses from earlier questions.

Recommendations/Reviews was a bit lower down the scale, floating in the area of 5th/6th overall.

1 11 1.9%
2 34 5.9%
3 49 8.5%
4 101 17.6%
5 146 25.4%
6 144 25%
7 90 15.7%

What I do find interesting is that while this factor was rated lower, there would seem to be a loose tie-in to the how much accuracy is valued. I’ve observed that our views on accuracy tend to be tied not to personally researched and verified data, but rather to data/opinions presented in reviews and recommendations. Of course, as I’ve noted, my observations are amatuer at best, so I’m not sure that much real meaning can be concluded, other than reviews aren’t of great importance overall.

Expected Build Experience was much lower than I thought it would be.

1 25 4.3%
2 31 5.4%
3 54 9.4%
4 80 13.9%
5 107 18.6%
6 158 27.5%
7 120 20.9%

Of course, that is probably due to my own biases regarding model building. I’m much more interested in the experience than I am in accuracy, for example. But the fact that this category was so close to the bottom did surprise me. The cynic in me concludes that as long as it is accurate and in the scale a modeler wants, who cares if one enjoys it? 🙂 

Price was the most oft picked answer for the last factor, and again, I was a bit surprised.

1 25 4.3%
2 47 8.2%
3 65 11.3%
4 83 14.4%
5 113 19.7%
6 83 14.4%
7 159 27.7%

Given the amount of discussion I’ve seen over the years regarding price, I had assumed it would have scored a bit higher. 

Then again, if all else is right- right subject, right scale, tolerable degree of accuracy, I suppose price tolerance is the most flexible, and therefore, most likely to be less of a factor.

Of course, it should be possible to do some math and determine exactly what order these average out to. But I’m not good at math, so if anyone would like to undertake that task, please contact me.

Do you participate in online scale modeling forums or other online modeling groups?

Not at all 38 6.6%
Occasionally 289 50.3%
Frequently 185 32.2%
Constantly 63 11%

I must admit, this one is a bit skewed, and I knew going in that it was. If you do not participate in some way with online modeling communities, you’d not have likely found out about this survey. So some of the answers have to be taken in light of that. The difference between “not at all” and “occasionally”, for example, can be quite subjective. If you don’t have much free time, spending 5 or 10 minutes a day may seem like “occasional”. But another person may classify that as “not at all”. So again, the numbers should be viewed with that reality in mind.

The largest group said they fell under occasional use, just over 50%. The next largest group- 32%, classified their use as frequent. Given the subjective nature of those terms, I think it is safe to say that a very large number of respondents spend a reasonable amount of time on the web in ways that relate to their hobby. (Disregarding reviews on accuracy for buying decisions seems to be a popular pastime while online…. I’m kidding, I’m kidding….)

Which single statement best reflects your attitude regarding the influence online communities have on your model purchasing decisions?

Zero Influence 77 13.4%
Some influence 385 67%
Significant influence 112 19.5%
Total influence 1 0.2%

This one digs a bit more into what people think of information gathered from the web when it comes to modeling decisions. But rather than seeing where it ranks regarding other items, this question focuses in on how much of an influence does the internet have on your buying decision specifically.

I was a bit surprised how large a number said “some influence”. Again, this is a bit subjective. But given how important accuracy is (from the questions above), I’d have expected at least a bit of an even split between “some influence” and “significant influence”. I say this because it would seem that the way a judgement of accuracy would be formed is through the internet, and given the desire (overall) for accuracy, the only way to determine that would be through online sources. And thus, it should follow, those online sources would have significant influence. 

Again, this is all very subjective, but I do find it interesting. In any case, it’s a fair statement to say that generally, online communities do play a part in purchasing decisions.

As an aside- and again I’m stirring the pot, I suppose- but it does make me reflect on how important accurate information is with regards to any particular model. As in almost any endeavor, the facts are often shouted down by a vocal minority. Yes, this is opinion, clearly. But it would seem it’s far more helpful to simply state facts than to shout opinion. I think that is why so many people trust Brett Green of Hyperscale, if I may single him out. I’ve always found his reviews and opinions to be very balanced and useful. (There are many others, of course, but Brett is, in my mind, the most outstanding example in our community.)

So keep that in mind. Be helpful. And if you can’t be helpful, be quiet. (Laugh, it’s funny….)

When you do participate in a model forum, what description best characterizes your involvement in the forum/online community?

I read only and never provide input 67 11.7%
I read and occasionally provide input 323 56.2%
I read and frequently provide input 13 2.3%
I read, provide input and occasionally start topics 151 26.3%
I read, provide input and frequently start topics 21 3.7%

This questions goal was to get some idea of how engaged the respondents were in online communities. The responses came out pretty much as I’d always suspected- there are a lot more people watching and listening than talking. But that’s not really surprising. Get any group of 10 people together at a coffee shop or in a home or at a cookout, and you’ll see that one or two dominate the conversation, one or two mostly listen, and the rest interject from time to time. I’d dare say the numbers were exactly the same when we were 7 years old and gathered on the baseball field or at a birthday party. And the personalities were essentially the same then too, I’d reckon.

So take heart… if you are generally a “lurker” on forums, you’re in the majority.

How often has the information presented in your favorite online community differed with your actual experience of building a model?

Every time 3 0.5%
Frequently 111 19.3%
Half the time 218 37.9%
Rarely 229 39.8%
Never 14 2.4%

The answers to this question leave us with a useful conclusion- in general, modelers find that when you read about someone’s experience building a model, it has a fairly high likelihood of describing what you’re own experience will be like. And this is no surprise, I suppose. I tend to read the build reviews of people who I’ve seen report a similar experience as I had. There is a level of trust there.

Of course, it can’t be ignored that a reasonably significant number- over 19%- said their experience differed frequently. And depending on your own personal viewpoint, those “half the times” could be lumped in with the “frequently” group.

In your *preferred* modeling genre and scale, what is the primary price range you look for? (In US$)

$25 or less 112 19.5%
$26-$50 281 48.9%
$51-$75 127 22.1%
$76-$100 39 6.8%
$100 or more 16 2.8%

This one surprised me a bit. I’d thought going in the numbers would have skewed a bit cheaper, based on anecdotal evidence. But this showed that modelers are actually a bit more price tolerant. Of course, a lot of this chart is driven by preferred scale, genre, and manufacturer, to a degree. 

The “sweet spot” though is definitely in the $26-$50 range, though with some tolerance for higher prices.

How many unbuilt models do you have in your possession? (Your “stash”)

Less than 10 27 4.7%
11-20 34 5.9%
21-50 71 12.3%
51-100 82 14.3%
100-250 145 25.2%
251-500 97 16.9%
501 or more 119 20.7%

I’ve received quite a few emails about this survey, and this question was, by far, the one everyone seemed most interested in. And I think the most common statement could be boiled down to a variation of “Am I crazy for having this many models?”, or, “Give me some numbers so I can show my wife I’m not that bad!” 🙂

Now, as we’ll see when we examine the numbers of built models, as a group, it would probably be more accurate to call us “collectors of models”.

I did some numbers, and I think this will blow your mind. (It did mine.)

If we assume that for each category, a reasonable number to do some projections with is the halfway point of any category, that would give us an inventory of over 105,000 models for the 575 respondents. That’s an average of 182 models per respondent. (It’s not a perfect number, of course, but it works for illustration.)

Now what blows me away is that I know, without a shadow of a doubt, that many of those who checked the 501 or more answer have collections that go far, far beyond that. I know folks who have 2500+ models, and I have heard of some from reliable sources that have well into 4000 and 5000 kits.

I’ll let each reader draw their own conclusions. But I do think it’s clear- there are a LOT of models out there sitting in closets, garages and so forth.

How many models have you *completed* the last 12 months?

None 72 12.5%
1 55 9.6%
2 80 13.9%
3 64 11.1%
4 84 14.6%
5 47 8.2%
6 43 7.5%
7 13 2.3%
8 19 3.3%
9 15 2.6%
10 21 3.7%
11-20 46 8%
21-40 9 1.6%
40+ 7 1.2%

This is where the rubber meets the road. We obviously build much, much less than we buy. (Manufacturers, take note…) 

Now, some of these numbers didn’t surprise me. I figured from anecdotal data that the likelihood for built models would be roughly in the 3-5 range. It registered slightly lower- 2 to 4 in the last year, but still about what I’d expect.

What I did not expect was that the 3rd largest group would be “none”. I think in a follow up survey, it would be interesting to explore how this relates to number of models purchased each year, as well as number in the stash. My observation has been that there is a relationship between large stash size and low finished kit output, but that is just my own observation of a few folks.

Of course, there are a few folks who built plenty of models in the last year. But they are the exception, really.

How many models have you *completed* in the last 5 years?

None 17 3%
1 10 1.7%
2 13 2.3%
3 22 3.8%
4 17 3%
5 35 6.1%
6 18 3.1%
7 17 3%
8 19 3.3%
9 5 0.9%
10 43 7.5%
11-20 152 26.4%
21-40 114 19.8%
40+ 93 16.2%

This question simply expanded the time frame from the previous question, and here we see the percentage of “None” dropped significantly. The math will basically confirm the annual averages to be roughly the same as the previous question.

Of course, in these two questions, I suppose health is a consideration also. I know several modelers who no longer build because of vision issues, mobility issues, various tremors, etc. Yet they do enjoy buying kits, talking about kits, etc. So I know from first hand evidence that the enjoyment of the hobby can be more than just building. Yet I also know that most of those folks would love to be able to build again.

How many models do you own that have been started, but have not been completed?

None 32 5.6%
1 44 7.7%
2 42 7.3%
3 58 10.1%
4 39 6.8%
5 55 9.6%
6 31 5.4%
7 20 3.5%
8 22 3.8%
9 3 0.5%
10 56 9.7%
11-20 85 14.8%
21-40 43 7.5%
40+ 45 7.8%

The old “shelf of shame”. Most of us have done this to some degree- you start a model, and then for some reason, you put it aside. It may sit for weeks, months, or years. I think the record I have personally seen was a friend who restarted (and finished!) a model after it sat in his closet for 36 years.

What I was not prepared for was the large numbers. Roughly one-third of all the modelers in the survey have 10 or more unfinished kits. And the number in the 40+ category really surprised me. (I wonder what the ceiling is on that one?)


So there you have it. A little glimpse, flawed as it is, into we as modelers. I am truly grateful to everyone who took the time to answer the survey, and especially to those who contacted me with encouragement.

I do plan to follow up on some of these questions over time, looking to dig deeper into some of the numbers, and to also do a better job of addressing the flaws that have emerged.

And finally, I’d like to answer some questions from a couple of folks who emailed me, who shall remain anonymous. They were not at all happy with this entire survey, or my handling of it, for some reason. Yet I’ve been in this hobby long enough to appreciate even the curmudgeons. So with a smile on my face, and an invitation to enjoy a cup of coffee with me anytime, I provide these answers:

  • I’m not quite sure what type of idiot I am, but I’d guess it’s one of the later Marques.
  • I’ve recently had a physical, and my doctor did verify that my head is nowhere near where you suggest. My wife, however, has made the same assertion you did, so I can’t rule out that sometime in the past, what you suggest may have, in fact, been correct.
  • Yes, actually, I do have better things to do. But for some reason, I chose this. You’re welcome big fella!

Thanks again everyone!


  1. Jon, A great article and hilarious to boot…enjoyed the process and the write-up. And regardless of what some of the curmudgeons might assert, you are not a later marque idiot, I mean at our age, that’s almost an insult!!

    1. Jon
      An interesting survey! As a 1/48 scale aircraft builder the results appear to mirror my own interest.
      One thing struck me as odd is the small number of people that responded to 1/35 scale (9%) but 51% build armor.
      One thing I would be interested in is the demographics of the responders.
      Overall, very well done!

  2. Jon, I enjoyed doing your survey and the results did surprise me. The price factor for example. Perhaps relate the price of a kit to the age of the modeler? overall I would say great job and I look forward to your next survey.

  3. Some excellent information. Some surprising results, some, not so much. The age breakdown is just downright scary. But then, I will be gone before that edge of the world falls off (I’m 68). Thanks for taking the time and effort expended in this survey.

  4. Although I generally recognize myself to be a curmudgeon (as does my wife), I thoroughly enjoyed your survey (I did participate), conclusions, and the comments. Will be looking forward to follow-ups.

  5. It would be interesting to present exactly the same survey to IPMS members where the results would be less skewed to aircraft modelers.

  6. Excellent article, and I found the results interesting; enjoyed reading thru the reduced data. Having said that I have a perverse fascination with statistical analysis, LOL

  7. 1,000,000 Green Monkey Mint Ninja Points for you !

    I actually think this is pretty accurate on all accounts. I know it is pretty close to the reality of my hobby world.

  8. I’m 32, so it’s kind of scary to think how few of us will be left in thirty to forty years. Kevin Johnson, Zachary Faull, and I may be doing the IPMS Nationals all by ourselves. 😳

  9. Jon,

    Thank you for doing this. My Top 10 questions, comments, and thoughts:

    1. I wonder how much the results were skewed by being posted on Hyperscale. As in HS is mostly 1/48 aircraft builders who have reputation for valuing accuracy. If the link had been posted at other locations, would it have influenced the outcome? I have to kick myself for not posting the link on the 72nd Forum. Didn’t even think about doing that until today.

    2. Where do you get off calling me middle age? 43 is young! Ok, maybe not. I was quite surprised at how low the the 19-40 representation was. Again, I wonder if a link posted on Facebook modeling groups would see a different result?

    3. I too am surprised…shocked even…how much price appears to be a non-factor in purchase. I find myself thinking twice about buying a kit that exceeds $25. But then I’m a 1/72 builder and cheap. I think this means that a moratorium should be places on all future discussions of price of model.

    4. I think your build experience would rank higher if I different poling location had been used…here I think Hyperscale skewed the results again. Raising accuracy and dropping build experience.

    5. Oh how I wish I could have answered less than 10 models in my unbuild collection. Oh, how I wish..

    6. I am not surprised with the yearly total. I’ve always suspected that a large portion of online modelers don’t finish kits. But maybe I’m just too cynical!

    7. I guess you are a finisher. How could not not have expected the large number of started but unfinished models? How, I ask, how?

    8. When you have seen Revell’s sales figures, you have to wonder who buys all the car models they sell? Are car modelers so fragmented that they keep themselves and won’t talk to us more warlike subject builders? Or is their market actually non-modelers. (And if so, how can they be non-modelers if they buy car models?)

    9. This was added just so I could have a Top 10.

    10. I am not Letterman, but I did attend a taping of his show once.

    1. Jim, thank you for an entertaining reply!

      I did post to a few other venues, including some Facebook groups, as well as other areas of Hyperscale. A few others reported they had posted it to other groups. To be honest, I went into this thinking there wouldn’t be much interest…. so I wasn’t prepared for success! (Lesson learned.) In future surveys, I will definitely reach out to a wider, more diverse audience.

      I finish 20-30 kits per year, and my stash is about 140. That number is going down though, as I focus more on building what is in the stash.

      Thanks again for your interest!

  10. Good survey. Most of the answers are about what I thought they would be. If you do decide to do this again (please) I’d suggest asking how long someone has been in the hobby, maybe their income level, & approximate % of that income spent on the hobby. Most of all, thanks

Leave a Reply

Your email address will not be published. Required fields are marked *