Correlational finding on Happiness and subject: Current contacts with friends

StudyArampatzi et al. (2018): study NL 2012
TitleSocial Network Sites, Individual Social Capital and Happiness.
SourceJournal of Happiness Studies, 2018, Vol. 19, 99 - 122.
URLhttps://link.springer.com/content/pdf/10.1007%2Fs10902-016-9808-z.pdf
Public15-44 aged, internet users, The Netherlands, 2012-2013
SampleProbability multistage stratified area sample
Non-Response
Respondents N =1944

Correlate
Author's labelContact with friends
Page in Source T1,3,5,AppendixA
Our classificationCurrent contacts with friends
Operationalization
Single question:
"How often do you do the following: Spend an evening 
with friends?"
7 Almost every day
6 Once or twice a week
5 A few times per month
4 About once per month
3 A number of times per year
2 About once per year
1 Never
Observed distributionRange:0-1; M;0.31; SD:0,46
Remarks
Order and hence sign of association reversed by 
WDH-team

Observed Relation with Happiness
Happiness
Measure
StatisticsElaboration/Remarks
O-HL-u-sq-n-11-jDM=- ns
                   Mean happiness
             Frequency of social media use/week
               <10 hrs     >10 hrs   DM
Few contacts    7.05        7.18     -.13
Many contacts   7.23        7.38     -.15 
- difference    +.18        +.20
O-HL-u-sq-n-11-jBeta=+.07 ns
Beta controlled for:
- time spent on social network sites
- time spent on internet (other)
- time spent on online games
- trust in people
- gender
- age
- employment status
- marital status
- education
- year dummies

Propensity score matching produces similar results
O-HL-u-sq-n-11-jBeta=-.04 ns
Beta additionally controlled for:
- infrequent contact with family 
- satisfaction with social contacts
- social loneliness


Appendix 1: Happiness measures used
CodeFull Text
O-HL-u-sq-n-11-jSelfreport on single question:

On the whole, how happy would you say you are?
10 totally happy
9
8
7
6
5
4
3
2
1
0 totally unhappy


Appendix 2: Statistics used
SymbolExplanation
BetaSTANDARDIZED REGRESSION COEFFICIENT by LEAST SQUARES (OLS)
Type: test statistic.

Measurement level: Correlates: all metric, Happiness: metric.
Range: [-1 ; +1]

Meaning:
beta > 0 a higher correlate level corresponds to a higher happiness rating on average.
beta < 0 a higher correlate level corresponds to a higher happiness rating on average.
beta = 0 no correlation.
beta = + 1 or -1 perfect correlation.
DMDIFFERENCE of MEANS
Type: descriptive statistic only.
Measurement level: Correlate: dichotomous, Happiness: metric
Range: depending on the happiness rating scale of the author; range symmetric about zero.

Meaning: the difference of the mean happiness, as measured on the author's rating scale, between the two correlate levels.
Source:
Ruut Veenhoven, World Database of Happiness, Collection of Correlational Findings, Erasmus University Rotterdam.
https://worlddatabaseofhappiness.eur.nl