Table 1.
Studies examining Similarity Effects of Self-Reported Big Five Personality Traits using APIM, and adding either Difference Scores, Profile Correlations, or Second-Order Polynomial Regression Coefficients
| Authors (Year) | Sample |
N couples |
Married | Age | Longitudinal design | Outcome | Similarity Measure | Similarity Effect?a |
|---|---|---|---|---|---|---|---|---|
| Dyrenforth et al. (2010) | HILDA (Australia) | 2,639 | 100% | Mmale = 50.96; Mfemale = 48.49 | no | GWB (1 item) | DS | yes, +extraversion (p < .05), and +neuroticism (p < .05) |
| GWB (1 item) | PC | no | ||||||
| RWB (1 item) | DS | yes, +extraversion (p <.01), +openness (p <.01), and +mean similarity (p < .01) | ||||||
| RWB (1 item) | PC | yes +PC (p < .01) | ||||||
| BHPS (UK) | 3,277 | 100% | Mmale = 51.67; Mfemale = 49.42 | no | GWB (1 item) | DS | yes, -agreeableness (p < .05), and +neuroticism (p < .01) | |
| GWB (1 item) | PC | no | ||||||
| RWB (1 item) | DS | yes, +neuroticism (p < .05) | ||||||
| RWB (1 item) | PC | no | ||||||
| GSOEP (Germany) | 5,709 | 100% | Mmale = 53.7; Mfemale = 51.0 | no | GWB (1 item) | DS | yes, +agreeableness (p < .01) | |
| GWB (1 item) | PC | no | ||||||
| Furler et al. (2013)b | SHP (Switzerland) | 1,608 | 100% | Mmale = 51.88, Mfemale= 49.10 | no | GWB (1 item) | DS | yes, -agreeableness (p < .01) |
| GWB (1 item) | PC | no | ||||||
| Hudson & Fraley (2014)c | Community sample (USA) | 174 | 0% | M = 20.37 | 5 time-points across 1 year | RWB (5 items) | DS | yes, +agreeableness (p < .05) |
| RWB (5 items) | Squared DS | yes, -neuroticism (p < .01) | ||||||
| Weidmann et al. (2017) | CoDip (Switzerland) | 237 | 70.9% | Mmale = 50.7; Mfemale = 48.4 | 2 time-points across 2 years | RWB (7 items) | PR + RSA | no |
| 141 | Change in RWB | PR + RSA | yes, +openness for women |
Note. We only reported studies that (1) controlled for actor and partner effects, and (2) controlled for the dyadic nature of the data (i.e., used an actor- partner- interdependence model). RWB = Relational Well-Being GWB = General Well-Being; DS = absolute Difference Score; PC = Profile Correlation, measured using standardized intra-class correlation across all Big Five traits; RSA = Polynomial Regression and Response Surface Analysis ; HILDA = Household Income and Labor Dynamics in Australia; BHPS = British Household Panel Study; HRS = Health and Retirement Study; GSOEP = German Socio-Economic Panel Study; SHP = Swiss Household Panel; CoDip = Co-Development in Personality Study.
A positive effect indicates that similarity is positively related to GWB or RWB.
In addition to absolute difference scores and standardized ICC, Furler et al. also tested the effect of variance similarity (i.e., absolute value of difference between the variances across all traits within the profiles) on general well-being, and found no significant effects.
The results represent effects on well-being across all time points. Hudson and Fraley also tested if people had higher relational well-being at time-points where they were more similar in personality, and found no significant effects.