Gallup’s Mood-Tracker
Check out the Gallup Daily US Mood Tracker:
http://www.gallup.com/poll/106915/Gallup-Daily-US-Mood.aspx
This chart comes from the Gallup-Healthways Well-Being Index, from a poll of Americans a day (claiming 98% coverage.) The survey contains questions about health, diet, well-being, stress, and economic indicators (“Although it’s not very likely that you did, could you tell me if you happened to purchase or lease a motor vehicle yesterday, such as a car, truck, or SUV?”) More info here: http://www.well-beingindex.com/. The mood tracking data is generated from the following question:
Q. Did you experience the following feelings during A LOT OF THE DAY yesterday? How about?
* Enjoyment
* Physical Pain
* Worry
* Sadness
* Stress
* Anger
* Happiness
There are two data series in the mood tracker, one for the percentage of respondents who experienced “a lot of happiness/enjoyment without a lot of stress/worry” (no indication if the / represents an AND or an OR) and another for the respondents who experienced stress/worry but not happiness/enjoyment. The two scores have an inverse correlation of -.82. There is a visible drop in the happiness index starting in early September with the stock market crash that is sustained until the present. The statistically significant drop reduces the average index score of 49.2 in January-August to 46.0 in September-Febuary.
hell yea!
But……how does data from a random sample of American households reflecting on their mood of the previous day compares to that from a heavily opt-in, self-selected group of HappyFactor users reporting their happiness moment by moment?
Here’s the intraweek comparison – with the Gallup data normalized to the HappyFactor mean.
What’s going on here? The HappyFactor intraweek data is pretty flat in comparison to Gallup. HappyFactor shows significant happiness bumps over the weekend, but nowhere near as large as Gallup.
First, maybe HF coverage bias…if HF users enjoy their work more than average Americans, or are spending their weekends drooling on the couch while the rest of the country engages in Bacchic hedonism…..yes, certainly possible. Other possibilities?
Well, we know that our memory of yesterday is blurred by the passage of time and influenced by our mood today. Suppose someone sitting at home at 8 p.m. on a Monday night picks up the phone and considers how she felt on Sunday. After a crappy day at work, she might report that Sunday was peachy indeed… even though her mood on Sunday was “actually” close to her mood on Monday.
Third, the internal stereotypes for what we are supposed to feel might also affect our memory of past moments differently than our real-time experience of those moments. By responding to a question about a certain day of the week (Yesterday…what day was that? Friday? Monday?) there could be activated representations (bear with me) of related concepts (I know). Case of the Mondays, TGIF. Those activations probably aren’t going to happen in a response to HappyFactor’s “how happy are you right now?”
Also, there’s always the issue of social desirability bias when you’re talking to other humans. Especially if she has the dulcet vocal chords of a trained sociologist…..perhaps with “extensive” public-sector experience at the Census Bureau….
…and skin like Neve Campbell….
Spicy.
This graph warrants a post by itself but it’s also interesting in the context of this discussion. It’s average HappyFactor by hour of the day. Besides the outlier at 8 AM (commute? Caffeine rush?) there are 3 pretty distinct clusters of happiness: morning to noon, noon to 5, and 5 to bedtime.
Yeah, the error bars are big, but isn’t this cool? Basically we’re miserable in the morning and stoked in the evening after work is over. This also reminds me of human cortisol levels, which are highest in the morning and drop throughout the day. I am grateful to work for Facebook, where I can skip out on some of those mood-raping morning hours!
If Gallup is always polling in the evening, they are actually sampling a much happier population than HappyFactor is. I’m not sure how that would muck with the data but it would be intriguing to dial people in the morning and what it looked like.
Hey, maybe you could predict the stock market….
