Are studies of ageing infected with ageism?

I have written many Newsletters about ageism.(See for example Newsletter #99 Age and ageism" How it impacts job opportunities, medical treatment and many areas of life. I turned recently to look at academic research. How ageism may be distorting the very results that I talk about.

Laboratory Experiments.

There are many studies done in the laboratories of psychology departments across the world. Experiments to look at the advertising are common. They ask things like “Are older people more sensitive to emotional advertising? Are older people less able to detect misinformation. Can older people be influenced by prescription drug advertising in the USA”.

The typical experiment creates two groups, one “old” and one “young”. They are shown some ads and asked to respond . The researcher then compares the scores for the groups and pronounces the results. There are multiple ageist biases that can and do creep in. Where do the samples come from? Don’t forget these laboratories are invariably on campuses. Students are recruited to be the “Young” group. Often, they receive extra course credit for participating. The “old” groups are also a “convenience sample”. They come from alumni, library members, campus or community groups for older people.

By using a convenience sample the researchers have adopted an ageist model. “All old people are the same”. Some studies provide a profile of their samples. The range of ages is 65 to 90. All are averaged out as “old”. We know that there will be declines in mental ability at some age. At 90 it may be present but at 66? Worse still the studies use different ages to define old. Some start as low as 50, others use 70. Many use 65 as some kind of “retirement” breakpoint. In this Newsletter we are learned that more and more people are not retiring then. In the UK state retirement age is 67.(See Newsletter # 111 Working Late) Most studies using longitudinal data show little cognitive decline until 80.

The researchers are just as ageist with the young group. They combine 18-year-old undergraduates with 26 year PHD students. They assume they are homogeneous. With such ageist based designs how much credence do we give to the results?

Ageism and Survey Research.

A lot of survey research takes age as an independent variable. It asks questions based on the format “ Does X increase as we get older?” Here we walk into a different ageism. This is another kind of homogeneous group problem. They assume that in fifty years time the forty year old in the sample will behave like the 90 year old. Unfortunately we know that there are generational effects. We are ageing more healthily in each generation. We only have to compare the health and behaviour of todays 75 year olds to those of a generation before. Our 40-year-old is on a different ageing “curve” to the ninety year old when they were 40

The Dangers of Averages

The Substack Community that looks at ageing is buzzing with a new study. It came from Becca Levy and was published this month. She is a famous researcher and has written extensively about ageism. It was her that theorized about the internal stereotypes we all hold. How they can become self-fulfilling. The article flags just how much ageing study conclusions are influenced by the stereotypes that we hold. Ageing is assumed to be about decline. After some pre-ordained age everything is decline.

Becca and her co-researcher set out to challenge this stereotype. The used a pre-existing longitudinal survey. From that survey they picked too measures. One assessed physical capability. The second was a standard measure of cognitive ability. They chose people over 65 who had been surveyed multiple times as part of the study. They started off by showing the dangers of averages. Across their large sample the average result showed a small decline in both measures year on year.

The problem with averages is that we do not know what they hide. Did everybody decline a little? Did most people stay the same and a few decline a lot? Where there a large group that declined and an only slightly smaller group that improved? There are many ways that the aggregate results could happen. Unfortunately decline is part of the stereotype. The average reader would assume that the vast majority of people declined and happily accept the results.( Did you?)

The disaggregated results are very different. They looked at the trajectory of individuals. A significant number of people showed an improvement. In fact 41% showed an improvement in either their physical or cognitive abilities. 32% improved in cognition and 28% in physical. Things got better after the age of 65. We can extend the logic by including those people whose scores did not change over the period. 51% of respondents were either stable or improved on cognition. 38% were stable or improved physically. These results covered some people up to the age of 90 and with up to 12 years’ worth of tracking.

These improvements were not marginal. Many people included in the stereotypical definition of old are not declining. Over half are not declining mentally.

Interpreting Ageing

We all hold stereotypes. It is part of the way our brain abstracts the multitude of information it receives. We have stereotypes of the citizens of a country or a particular town. We stereotype by race, religion and sex. If those stereotypes acquire negative connotations, we become prejudiced and discriminatory. That can influence something as apparently objective as academic work. Interpreting and generalizing across ageing studies we need to watch out for our own and the researcher’s ageism. Just because the results conform to our prejudices does not mean that they are correct. Just because a study makes it to the mainstream media does not mean it is well researched.