Prediction is difficult, especially when it involves the future

In late 1949, a Gallup Poll asked people about their predictions for the next 50 years.

IN JUST A FEW WEEKS IT WILL BE 1950 - AND HALF OF THE 20TH
CENTURY WILL BE OVER. NOW, I'D LIKE TO ASK YOU SOME
QUESTIONS ABOUT THE NEXT 50 YEARS:

Do you think there will be another world war during the next 50 years, or not?
Yes—81%, no—11%, don't know—7%

Do you think a cure for cancer will be found?
Yes—89%, no—7%, don't know—4%

Do you think trains and airplanes will be run with atomic power?
Yes—64%, no—21%, don't know—15%

Do you think men in rockets will be able to reach the moon or the planet Mars?
Yes—16%, no—69%, don't know—14%

Do you think people in this country will go to church more often or less often than they do now?
More—40%, less—31%, same—22%, don't know—7%

Do you think labor unions in this country will become stronger or weaker than they are now?
Stronger—45%, weaker—33%, same—15%, don't know—8%

Another Gallup Poll conducted at the same time asked about some other possibilities:

Do you think the number of working hours per week for the average person will be cut down to about 30 hours—or do you think that the number of hours will stay about where it is today?
Cut—47%, same—39%, don't know—14%

During the next 50 years, do you think that most of the nations of the world will have a democratic government like the U. S., a Communistic government like Russia, or a socialistic government like England?
Democratic—54%, communistic—7%, socialistic—11%, other—3%, “no code or no data”--24%

Do you think a woman will be elected President of the United States at any time during the next 50 years?
Yes—31%, No—59%, Don't know—9%

During the next 50 years, do you think that the government in Washington will own and run  …. the railroads in the U. S.?
Yes—37%, No—43%, Don't know—20%

...the banks.
Yes—28%, No—53%, Don't know—18%

….the big industries, like auto, steel, etc.
Yes—24%, No—57%, Don't know—18%


By my count, that means the majority was right on six and wrong on six, which doesn't seem too bad. 

PS:  The "reach the moon or the planet Mars" is a classic example of poor question design.  I'm surprised they didn't notice that.    On the prevailing form of government, they didn't show a "don't know" category.  I'd guess that "no code or no data" is mostly people who said they didn't know. 

America and Britain

In 1973, the United States Information Agency sponsored a survey of British elites concerning their views of the United States and Britain. It included seven types of elites: government, arts, academics, media, student organizations, business, and labor. I thought that there might be large differences among the groups—for example, business would be more favorable to the United States, arts or academics would be more critical. But although there were some differences, they weren't very striking. The comparison of general views of the nations was more interesting. The survey asked how successful each nation was in providing different things, on a scale ranging from 1 (“very successful” to “not successful at all”). The differences between the average ratings, with positive numbers meaning that Britain was seen as doing better than the United States:

1.51 'taking care of its sick people'
1.24 'rights of its minorities'
1.06 'minimizing drug addiction'
1.03 'peaceful means for changing social conditions'
1.02 'taking care of its poor people'
1.22 'maintaining law and order' 
  .50 'taking care of its old people'
  .35 'good quality education' 
  .35 'minimizing environmental pollution'
  .24 'jobs for everyone who wants to work'
  .18 'full educational opportunities for most people'
  .02 'adequate standard of living for most people'
  .01 'encouraging full development of the arts'

-.32 'stable economy'
-.66 'opportunites for personal career advancement'

It also asked how well certain pairs of terms described British and American societies:
for example, “class conscious” versus “classless,” with the scale ranging from 1 to 7.
The ratings:

2.81 'non-violent'
2.48 'stable'
2.44 'cooperative'
2.22 'law-abiding'
1.86 'disciplined'
1.72 'tolerant'
1.43 'non-materialistic'
.66 'permissive'

   .00 'non-conformist'
-0.10 'innovative (vs. imitative)'
-0.18 'ideological (vs. pragmatic)'
-1.42 'classless'
-2.52 'dynamic'


For example, British elites saw the term “non-violent” as more descriptive of Britain than the United States, and “dynamic” as less descriptive of Britain than the United States. The words in parentheses are the opposing term when it isn't obvious.

Christmas shopping guide (slightly outdated)

In December 1958, a Gallup Poll asked "If you could have your choice, what one present would you most like to have for Christmas?"  About 14% didn't know, said that there was nothing, or didn't answer.  About 7% named some kind of general improvement in their living standards ("lots of money," "a job"), and about 6% said a new house or apartment. Other popular choices: 

New car                                       14%

Clothing                                      11%

Large appliances (e. g., a stove)              8%

Electronics                                    7%

Furniture, home furnishings                    6%

Sporting/Hobby equipment                       4%

Jewelery                                       2%

Small appliances (e. g., sewing machine)       2%

Vacation, trip                                 2%

What struck me was how practical most people were.  Even most of the things I've counted as sporting/hobby equipment might have some practical value--the group includes 27 people who said "tools," compared to two who said golf clubs.  "Jewelery" includes 24 who wanted a watch, compared to only two who wanted a bracelet, and one who wanted pearls.  Of course, there were a few dreamers:  one person want an airplane. 

People were also remarkably centered on their homes.  No one mentioned anything related to dining out or entertainment, except for five who said "a good meal."

This question, or similar ones, have been asked a few times since then, but I haven't looked at them in enough detail to compare them. 

Of a certain age

A 1958 Gallup poll asked  "at what age . . . a woman reaches the peak of her beauty."  Answers ranged from 15 to 75, with an average of 29.9.  Yes, but there was a big difference between the opinions of men and women, right?  Not really--women gave an average answer of 30.1 and men gave an average of 29.6.  The difference is not statistically significant. 

It also asked "at what age does a person reach the peak of his mental ability"?  Answers ranged from one to 83, with a mean of 39.4.  Here, there was a difference between men and women--women gave an average of 37.8, and men an average of 41.1. When I was younger I might have been able to think of a plausible explanation for this difference, but now I'm coming up blank.

There were a few other questions about ages:  when middle age begins (average of 43.9 years), when old age begins (65.3), when an unmarried man should be described as a bachelor (33.8), and when an unmarried woman should be described as a spinster (32.4).  For all of them, women chose a slightly (but significantly) higher average. 

They do things differently there

From a 1959 French survey:

"When you were, let's say between 8 and 12 years of age what did you usually drink at meals?"

Only water            22%

wine with water added 44%

a little wine, pure    4%

cider                 15%

beer                  10%

other                  4%

There was a series of questions about the first time you had "the impression that you had drank a little too much."  One of them asked "Was it at a particular occasion and which one?"

15% Ceremonies of a family character very often religious followed by celebrations 
 8% Gatherings [at a] family meal 
11% Gathering [of] friends
 7% Popular celebrations
 8% Usual drinking excess marking the beginning of adult life
 3% During work
 2% At the army
 4% Curiosity or childhood feat unknown to the adults
 4% Accidentally, exceptionally, taken surprise by circumstances
 2% Other answers
36% Don't know or don't remember


"How did your parents react when they observed that you had effectively drunk a little too much for the first time?  Were they..."

amused                        11%
kind                          16%
did they think it was normal   4%
judged you severely            4%
they didn't know about it     20%
No answer                     40%


The question about the occasion on which you first drank a little too much was open-ended:  people gave whatever answer they wanted and later someone classified it into the categories.  The report gave some examples of people's answers:

"It's sad to say but it was my first communion"
"pig fair"
"when I buried my life as a boy"
"I was an apprentice for an old gardener who made me drink"
"after school, to celebrate his little brother's birth, a friend took me to a cafe"
"I had gone to be paid by a wine-grower and they made me taste their wine"
"to be a braggart"
"I was coming back from a long stay in England where I had drunk only water"
"it was at boarding school, we were well taken care of"

Will this reading be on the test?

My friend and colleague Brad Wright has included several plugs for this blog on his own blog.  Brad is interested in religion, so here's a finding for him, from a 1965 Gallup Poll.  The survey asked people whether they'd done various things--there was no particular logic to the list, which included flying in a plane, eating caviar, and betting money at a race track (more common than I would have guessed).  One of the items was "read the Bible through--every word."  20% of the respondents said they had.  The straightforward interpretation is that 20% of Americans had read the Bible all the way through, but I find that hard to believe.  The Bible is a long book, and as a college professor I have a good sense of the chance that people will read long books all the way through.

Another possibility is "social desirability bias"--people are giving the answer that they think they "ought" to give, rather than telling the truth.  If the question involved general frequency of reading the Bible or attending religious services, social desirability might be an important factor.  But I don't think many people imagine that the world will think worse of them if they skipped a few of the minor prophets.  Also, the way that the question was asked cut down on the risk--it gave them a list and asked them to read off the letters designating the things they had done.  So you didn't have to explicitly say "no"--you just had to leave the letter "p" off of your list. 

A third  possibility, which I find the most plausible, is that most of these people thought they had read the Bible all the way through, even though they probably hadn't actually done so.  How would this be possible?  My guess is that many people who attended church frequently, or had done so at some point in their lives, figured that given the Bible reading involved they must have covered it all over the course of time, even if not in consecutive order. 

Another question in the list asked if you'd read a book all the way through since leaving school.  Of the people who said they hadn't, 12% said they'd read every word of the Bible.  People who'd gone to college were most likely to say that they'd read the entire Bible (26%), but people who'd just reached grades 8-11 were almost as likely (22%).  Among high school graduates who hadn't attended college, it was only 15%.

Are we downhearted?

A CNN/Opinion Research Corporation poll question:  "As you may know, the United States went through a depression in the 1930s in which roughly one out of four workers were unemployed, banks failed across the country, and millions of ordinary Americans were temporarily homeless or unable to feed their families. Do you think it is very likely, somewhat likely, not very likely, or not likely at all that another depression like that will occur in the US within the next 12 months?"

                       Oct 2008  Oct 2008   Dec 2008   Mar 2009  July 2009  Dec 2009
Very            21%      12%     10%     12%      12%     17%
Somewhat        38%      29%     28%     33%      29%     26%
Not very        29%      35%     41%     36%      36%     39%
Not likely      13%      24%     20%     18%      22%     19%

There was a substantial change between the first two surveys, which were only two weeks apart (Oct 3-5 and 17-19), but very little change since then.  I'm not even sure if the differences among the last five surveys are statistically significant.  (If you want to calculate it yourself, the sample sizes are about 1,000 for each survey).  After a year when the worst didn't happen, it would be reasonable to think that it wasn't going to happen in the next year.  But that wasn't how people reacted.  

The improvement over a two-week period in October 2008 is also interesting.  The TARP bill was signed into law on Oct 3.  Maybe the process leading up to that contributed to a sense of panic, and people just bounced back after a few weeks.  Or maybe something positive happened between early and mid-October that affected people's outlook, although I don't remember the history well enough to think of what that might be.