Een -3.29 and 3.29. We then conducted a piecewise regression to examine whether the change ellbeing association differed between two discrete categories: Monocrotaline web people who experienced one standard deviation or less of positive trait change, and people who experienced more than one standard deviation of trait change. We refer to these categories as moderate and large changers respectively. (The first category included all participants with a standardized change of score of 1 or less–participants with negative trait change or stability were included, bmjopen-2015-010112 and thus the term moderate is only used for convenience.) In the piecewise regression, which was conducted in Stata, we regressed well-being at time 2 on the trait change jir.2014.0227 score, controlling for the same covariates as in the main analysis, and computed the regression slope separately for the moderate and large categories. In Table 6, these slopes are displayed in the rows labeled moderate and large respectively. The associated scatterplots and lines of best fit are displayed in Fig 8, where the contrast between the slopes for moderate and large changes is evident. (The scatterplots are bivariate plots without adjustment for covariates.) The full piecewise regression results can be found in a supporting file (S2 File). The row labeled Slope displays the difference between these two slopes, and the row labeled Interval displays the distance between the predicted well-being score of a person with a standardized change score of 1 and person with a standardized changePLOS ONE | DOI:10.1371/journal.pone.0131316 July 10,22 /Investigating the Goldilocks HypothesisTable 6. Trait Change Analyses: Piecewise Regression with Moderate Changers and Large Changers. EWB Wave 2 B Sociality Moderate Large Slope Interval Agency Moderate Large Slope Interval Conscientiousness Moderate Large Slope Interval Neuroticism (Rev.) Moderate Large Slope Interval 0.18*** 0.21** 0.03 -0.12 0.02 0.07 0.07 0.07 0.30*** 0.NA Wave 2 (Isovaleryl-Val-Val-Sta-Ala-Sta-OH cost reversed) SEB 0.02 0.08 0.08 0.07 0.02 0.08 0.08 0.06 0.02 0.08 0.08 0.08 B SEB 0.03 0.11 0.12 0.10 0.03 0.11 0.12 0.09 0.03 0.11 0.12 0.11 0.03 0.10 0.10 0.09 BPWB Wave 2 SE B 0.02 0.09 0.09 0.07 0.03 0.08 0.09 0.07 0.03 0.09 0.09 0.09 0.03 0.07 0.07 0.0.23*** 0.06 -0.17* -0.12 0.10*** 0.01 -0.12 -0.03 0.08*** -0.05 -0.13 0.0.23*** -0.01 -0.24* -0.07 0.13** -0.08 -0.20 -0.03 0.13*** 0.10 -0.03 -0.0.22*** 0.01 -0.20* -0.11 0.12*** 0.04 -0.08 -0.06 0.08** -0.18* -0.27** -0.05 0.15*** 0.10 -0.06 -0.-0.12 -0.24*Note. Absence of NA was square-root transformed prior to analysis in order to minimize skewness. p < .10 * p < .05 ** p < .01 *** p < .001. doi:10.1371/journal.pone.0131316.tscore marginally below 1. Negative values here indicate that the person with the marginally lower trait-change score had a higher predicted well-being score. For three traits--sociality, agency, and conscientiousness--the piecewise results support the argument that positive change confers benefits up to an inflection point. For moderate changers--those below the Goldilocks point--the slopes were positive and statistically significant in all nine analyses, indicating that incremental increases in positive trait strength (up to this point) predicted incremental increases in well-being. But for large changers, the slopes in eight out of nine cases did not differ from zero. In five of the nine analyses, the slopes in the moderate range and large range were significantly different from each other, but the non-significance in the other.Een -3.29 and 3.29. We then conducted a piecewise regression to examine whether the change ellbeing association differed between two discrete categories: people who experienced one standard deviation or less of positive trait change, and people who experienced more than one standard deviation of trait change. We refer to these categories as moderate and large changers respectively. (The first category included all participants with a standardized change of score of 1 or less--participants with negative trait change or stability were included, bmjopen-2015-010112 and thus the term moderate is only used for convenience.) In the piecewise regression, which was conducted in Stata, we regressed well-being at time 2 on the trait change jir.2014.0227 score, controlling for the same covariates as in the main analysis, and computed the regression slope separately for the moderate and large categories. In Table 6, these slopes are displayed in the rows labeled moderate and large respectively. The associated scatterplots and lines of best fit are displayed in Fig 8, where the contrast between the slopes for moderate and large changes is evident. (The scatterplots are bivariate plots without adjustment for covariates.) The full piecewise regression results can be found in a supporting file (S2 File). The row labeled Slope displays the difference between these two slopes, and the row labeled Interval displays the distance between the predicted well-being score of a person with a standardized change score of 1 and person with a standardized changePLOS ONE | DOI:10.1371/journal.pone.0131316 July 10,22 /Investigating the Goldilocks HypothesisTable 6. Trait Change Analyses: Piecewise Regression with Moderate Changers and Large Changers. EWB Wave 2 B Sociality Moderate Large Slope Interval Agency Moderate Large Slope Interval Conscientiousness Moderate Large Slope Interval Neuroticism (Rev.) Moderate Large Slope Interval 0.18*** 0.21** 0.03 -0.12 0.02 0.07 0.07 0.07 0.30*** 0.NA Wave 2 (reversed) SEB 0.02 0.08 0.08 0.07 0.02 0.08 0.08 0.06 0.02 0.08 0.08 0.08 B SEB 0.03 0.11 0.12 0.10 0.03 0.11 0.12 0.09 0.03 0.11 0.12 0.11 0.03 0.10 0.10 0.09 BPWB Wave 2 SE B 0.02 0.09 0.09 0.07 0.03 0.08 0.09 0.07 0.03 0.09 0.09 0.09 0.03 0.07 0.07 0.0.23*** 0.06 -0.17* -0.12 0.10*** 0.01 -0.12 -0.03 0.08*** -0.05 -0.13 0.0.23*** -0.01 -0.24* -0.07 0.13** -0.08 -0.20 -0.03 0.13*** 0.10 -0.03 -0.0.22*** 0.01 -0.20* -0.11 0.12*** 0.04 -0.08 -0.06 0.08** -0.18* -0.27** -0.05 0.15*** 0.10 -0.06 -0.-0.12 -0.24*Note. Absence of NA was square-root transformed prior to analysis in order to minimize skewness. p < .10 * p < .05 ** p < .01 *** p < .001. doi:10.1371/journal.pone.0131316.tscore marginally below 1. Negative values here indicate that the person with the marginally lower trait-change score had a higher predicted well-being score. For three traits--sociality, agency, and conscientiousness--the piecewise results support the argument that positive change confers benefits up to an inflection point. For moderate changers--those below the Goldilocks point--the slopes were positive and statistically significant in all nine analyses, indicating that incremental increases in positive trait strength (up to this point) predicted incremental increases in well-being. But for large changers, the slopes in eight out of nine cases did not differ from zero. In five of the nine analyses, the slopes in the moderate range and large range were significantly different from each other, but the non-significance in the other.
