In defense of the term "Millennial"

There's room between "Millennials aren't real" and "LOL Millennials love skinny jeans".

Reader note: this weekend, I’ll have Michael Dimock, president of the Pew Research Center, on my SiriusXM Show (“The Trendline with Kristen Soltis Anderson” is on P.O.TU.S., Channel 124, starting 10 am Saturday) to discuss Pew’s new approach to generational analysis. We’ll have a spirited debate! Please tune in.

For whatever reason - perhaps, just spitballing here, my extensive digital footprint writing about Millennials for the last decade and a half and a book literally titled The Selfie Vote: Where Millennials Are Leading America And How Republicans Can Keep Up - Instagram is really convinced that I want to see a steady stream of content about the differences between Millennials and Gen Z.

Here’s an example of the genre, from creator Cruz Corral, who makes videos about how the different generations think about remote work.

I have a lot of Gen Z folks at my company who are, of course, excellent and nothing like the caricature here. It turns out stereotypes are not perfect and not applicable to everyone in a particular demographic group, who knew! At the same time, there are moments like the Millennial character muttering “why is there so much text on that slide?” that hit all too close to home. These videos are amusing and, in my view, pretty harmless.

There are plenty of experts (and “experts”) out there who will deploy these generations to make generalizations, which can be annoying but ultimately isn’t the end of the world. “Millennials love skinny jeans, Gen Z hates them!” is not the end of the world.

Photo by Maude Frédérique Lavoie on Unsplash

But let me tell you, there are quite a few people out there who get really spun up about the use of generational labels like “Millennial”. Even Millennials themselves didn’t warm to the term. Whenever I’ve heard pushback about using a term like “Millennial”, I’ve always defaulted to an appeal to authority: The Pew Research Center says generations are real and here’s how they define them.

So you can imagine my surprise and sadness at learning that Pew has decided to largely dispense with generational analysis moving forward.

Last week, the team at Pew released a report explaining their reasoning behind it, which can be summarized as:

By choosing not to use the standard generational labels when they’re not appropriate, we can avoid reinforcing harmful stereotypes or oversimplifying people’s complex lived experiences.

With these considerations in mind, our audiences should not expect to see a lot of new research coming out of Pew Research Center that uses the generational lens.

When I read their write-up of the thought process behind the change, I was left unsatisfied - though not necessarily because I disagree with their methodological concerns about the way generations get used (and abused) as a frame for analyzing public opinion. Instead, I felt like their proposed approach goes too far toward letting the perfect be the enemy of the good.

I know a lot of my reaction may be driven, deep down, by my own selfish desire to have their amazing research as a resource on this topic moving forward, but I also think it’s worth defending the use of a term like “Millennial”.

Let’s unpack why I think Pew’s heart is in the right place, but why I still don’t agree with their conclusion to largely stop analyzing groups like Millennials.

First, Pew says that “We’ll only do generational analysis when we have historical data that allows us to compare generations at similar stages of life.” This means that they won’t use generational analysis to compare young people today to old people today. (They’ll still make comparisons, they just will use age categories rather than generational labels.) Instead, they will only use generation labels if they can prove that, say young people are different now compared to young people in a decade past.

The rationale behind this is sound: just because something is a young people thing does not mean it is necessarily a Gen Z thing. For something to be unique to a generation, it needs to have a level of stickiness to it, to be something that is a defining characteristic of people not because they are 30-39 years old today but rather because they were born between 1984 and 1993. The only way to really, firmly know that is a generation thing and not just a young people are always like this thing is to compare apples to apples over time.

The problem, as Pew notes, is we don’t have apples to apples over time because the way a 25-year-old would have been polled in 1990 (probably over the phone) is not the same a the way a 25-year-old today would be polled (probably online). And changes in methods can have a real effect; just take a peek at my colleague

Patrick Ruffini

at his newsletter

The Intersection

highlighting how a viral poll finding about Americans supposedly becoming less patriotic can be explained by changes in how the questions were presented to respondents in 1998 versus today.

At the same time, I wish Pew would give a few examples of why they think comparing a survey of young people from 1993 done over the phone to a survey of young people in 2023 done online is so totally flawed as to be completely useless. Can’t we say, yes, the method changed? We don’t know how much of the difference in view is attributable to the change in method versus actual change over time, though we believe it is at least a partial contributor, so please take this trendline with an appropriate grain of salt?

If we are throwing every pre-online poll into the trash bin and throwing up our hands, that just feels like a terrible waste of potentially really interesting data. And if all we have to rely on, say, is stuff that the U.S. Census Bureau has been asking in the same way for the last few decades, isn’t that limiting and fraught in and of itself?

Next, Pew stays “Even when we have historical data, we will attempt to control for other factors beyond age in making generational comparisons.” I don’t have any objection to that methodologically. In fact, I make the case in my own book that you can’t separate out some of the big demographic differences present in younger generations from the analysis of the generation itself.

Quoth The Selfie Vote:

The reality is that America’s growing diversity is intertwined with the new attitudes of the younger generation, and this fact will only become more apparent with each passing election.

It is absolutely the case that younger Americans are more racially and ethnically diverse, more comfortable with technology, less religious, more highly-educated, and on and on relative to older Americans. And so I understand the instinct to say, well, the difference between Millennials and Boomers might just be because Millennials are more likely to have a graduate degree, so what’s really going on here is about education, not generation, and so on.

I understand wanting to know what demographic factor is driving a view and which are incidental. But at a certain point, I just don’t see the real-world value in saying “well, imagine a world where Millennials were actually less educated and less racially and ethnically diverse, only then can we actually compare them to the Boomers.” This is the rabbit hole that never really ends. We don’t live in a real world that “controls for variables”. We live in a messy reality where people wear a lot of different hats and labels that all intersect in interesting ways and shape our worldviews.

But let’s assume Pew gets data that checks those prior boxes. They then say: “When we do have the data to study groups of similarly aged people over time, we won’t always default to using the standard generational definitions and labels.”

Let’s face it: at a certain level, a word like “Millennial” is a made-up term that grew out of a basic human desire to put labels on things and to group people together to understand them better. As an “elder Millennial” or “geriatric Millennial”, I probably share more common cultural experiences and memories of social and political change with a young Gen X-er than I do someone born in 1995.

This is a big piece of Pew’s rationale for walking away from a word like “Millennial”. As they put it:

Existing generational definitions also may be too broad and arbitrary to capture differences that exist among narrower cohorts. A typical generation spans 15 to 18 years. As many critics of generational research point out, there is great diversity of thought, experience and behavior within generations.

So if Pew wants to jettison words like “Millennial” to instead look at narrower, more creative or topic-specific division points - like the first election in which someone was eligible to vote, or their age during the pandemic, or what have you, that’s fine. I will probably celebrate it!

But opinion researchers regularly offer analysis based on broad categories, like gender, race, or party. Sometimes people will say, well yes, but “women are not a monolith”, “Latino voters are not a monolith”, and yes, obviously that’s true. Researchers who refer to demographic breaks in their data are not trying to claim that all women or all Hispanic voters or all Democrats are alike; indeed, are we not doing the exact opposite when we say “70% of women believe X” - a statement that naturally means 30% do not, and therefore there is not unanimity in the group?

In my view, using a term like “Millennial” is basically no different than using other broad categories to explain our data. Pointing out differences between groups can always lead to stereotyping, always risks overgeneralizing, but we do it anyway, even as we say “groups are not monoliths”, because it helps us understand the world around us a little better.

You can always slice and dice further, of course, looking at smaller and smaller subgroups, but eventually it all dissolves to nothing, to "“not all women aged 30-39 living in the mid-Atlantic who have golden retrievers are a monolith!” So it doesn’t make a ton of sense to say a category like “Millennial” is too broad but then to still do analyses of even larger, broader subgroups. The only way in which it could make sense is to say that gender, or race, or age are basically fixed, real things, while “Millennial” is a made-up term.

But I do think that a grouping like “Millennial” isn’t totally arbitrary. Pew put a lot of thought into coming up with the dividing lines back when they did! Someone who was born in 1979 probably made it most of the way through high school before they got an email address or a cell phone. The way the explosion of people’s use of internet dramatically disrupted the world around the dawn of the millennium means that yes, in just a very short time frame, people on one side of the technological line had a very different youth than those on the other side of the line. (That’s why before we were called Millennials we were called “digital natives”.)

To me, the dividing lines between Gen X and Millennials and Gen Z aren’t actually that silly, because if you were someone born in 1996 (and there has to be a cutoff somewhere) not only was the internet always “there” during your adolescence and early adulthood, but so too were smartphones and social media. I think a lot of the things that people notice as differences between Millennials and Gen Z these days aren’t just perennial “30-somethings are always annoyed with 20-somethings, 20-somethings always think 30-somethings are uncool” type differences, but are really rooted in the extremely rapidly changing world we all now inhabit and the changing technology that has shaped that world.

The term “Millennial” can be a broad, made-up term but also have real meaning and utility in describing the world around us, both where we are and where we are going.

For instance, just earlier today, Nate Cohn at The Upshot has a piece up with a headline “Millennials Are Not an Exception. They’ve Moved to the Right.” This is the kind of assertion that has me calling the banners and preparing for war. (Just kidding! But watch this space!) People are still going to talk about Millennials even if Pew doesn’t. Yes, these words are charged and clicky, but I know I’m not going to stop getting asked about Millennials and Gen Z, and I don’t think people are wrong for asking about Millennials and Gen Z, so I’m not going to stop using the term where it makes sense. There’s a lot of room for “when it makes sense” that, for me, is probably short of Pew’s new very high bar but is still worlds away from “Boomer moms don’t understand almond milk.”

The bright folks at Pew have a lot of good reasons for treading carefully when it comes to generational analysis. I’ll miss what they have to say, and they’re not wrong that terms like “Millennial” get deployed to push pseudoscientific marketing gobbledygook that folks like us then have to go in and debunk.

However, I think the decision to walk away from most generational analysis in the face of criticism goes a step too far, and cedes the terrain to those very viral videos and marketing gobbledygook they decry.