What we've found
If you feel that alcohol is having a bad effect on your life, do contact your GP. There is lots of help available. Drinking more than 14 units per week can have significant impact on your health and drinking more than 6 units in one session is classed as binge drinking.
These insights are based on correlations which are statistical links between two sets of data. While we focus only on more plausible and interesting links, they should not be seen to imply cause or effect.
Is alcohol linked to happiness?
Alcohol may ‘take the edge off’ and help us feel a little more relaxed, but too much too often can have negative effects on our overall wellbeing and physical health. Therefore, we wanted to explore how alcohol consumption may be linked to happiness levels among our community.
A small amount of alcohol per week seems to increase the likelihood of reporting a high happiness score. However, when alcohol consumption exceeds the NHS guidelines of 14 units per week, users seem much less likely to report feeling happy.†
Can we learn anything from those in our community who avoid binge drinking?
We can see from the graph, that users who drink less than 6 units in one session are more likely to fall asleep quickly and less likely to smoke. However, we can't say that one causes the other.‡
If you’re looking for ways to reduce your alcohol consumption, take a look at our 8 simple steps to cut down.
Correlation ≠ Causality:
All statistical associations mentioned on this page represent statistically significant (95% level) correlations only. We do not make any claim of causality running in either direction between any two variables discussed, and have controlled for variation in age, sex and the Index of Multiple Deprivation (IMD) decile only.
Estimated alcohol consumption and happiness:
This graph uses data from over 17,800 Evergreen Users who have answered a question about happiness, the amount of units they typically drink, and the frequency in which they drink. Weekly consumption is estimated by taking the mid-points of the frequency and unit response categories, multiplying, and scaling to weekly. A locally weighted scatterplot smoothing method is then used (see https://bit.ly/3pcmdMd) to generate the curve and its confidence intervals (95% Confidence level). All consumption values, including zero, are used to estimate the curve.
The displayed correlations are the coefficients from 6 linear probability models, one for each outcome on the left, which control for drinking fewer than 6 units in a session, age, sex, and IMD decile. Smoking behaviour pertains to the previous 30 days and being anxious is defined as having a score greater than or equal to 7.