Situation Characteristics Are Age Graded

Abstract

The current work examines mean-level patterns of major dimensions of situation characteristics—the Situational Eight DIAMONDS—across the life span. Using population-representative data from the 2013 and 2014 American Time Use Survey (Study 1) and the 2012 German Socioeconomic Panel (Study 2), we tested hypotheses generated from research on situation cues and personality development. Results demonstrated that the DIAMONDS characteristics were age graded: Individuals tended to be in different kinds of situations as a function of their age. Furthermore, there was evidence that some patterns were country specific, whereas others replicated across the United States and Germany. Overall, these studies suggest that—much like personality traits—situation characteristics have predictable mean-level patterns over the life span.

Keywords

situations Situational Eight DIAMONDS Day Reconstruction Method life span mean-level patterns

Many developmental perspectives on human functioning, personality stability, and interpersonal processes concern how people’s environments unfold across the life span. Age-graded differences in situations could so far not be examined within a coherent situational taxonomy. Roberts and Pomerantz (2004, p. 402) were left to conclude that “[u]nfortunately, the assessment of environments has not been as thorough or systematic as that of psychological attributes, so we do not have the opportunity to catalogue all forms of consistency and change for even a subset of situations or environments.” However, there has recently been a resurgence in psychological situation research (Reis, 2008) with renewed interest in how to describe, taxonomize, and measure situational information (Rauthmann, Sherman, & Funder, 2015a, 2015b). Thus, this work is able to examine age differences in mean-level patterns of major dimensions of situation characteristics in representative U.S. and German samples, covering together an age range from 15 to 95 years.

Background

Describing and Measuring Situations

Rauthmann and colleagues (2014) as well as Rauthmann, Sherman, and Funder (2015a, 2015b) have proposed to focus on psychological characteristics of situations rather than raw, objective stimuli (cues). These can be used to efficiently describe any situation (de Raad, 2004; Edwards & Templeton, 2005), similar to how persons can be succinctly described by traits. They summarize the psychological meaning of a situation (e.g., work has to be done) instead of listing single cues (e.g., four books and three people) or categorizing entire situations into nominal classes (e.g., a work situation). As continuous characteristics grant a variable-oriented, dimensional approach to studying what situations mean, they are a very useful unit of analysis for situation research (Rauthmann et al., 2015a).

Rauthmann and colleagues (2014) suggested, based on extensive analyses across different countries, that situation characteristics may be captured in a parsimonious taxonomy: the Situational Eight DIAMONDS (Duty: Does work need to be done? Intellect: Is deep thinking processing required? Adversity: Is someone threatened? Mating: Is the situation sexually and/or romantically charged? pOsitivity: Is the situation pleasant and enjoyable? Negativity: Could negative feelings ensue? Deception: Is trust or mistrust an issue? And Sociality: Can meaningful social interaction and relationships develop?). The authors showed that these eight dimensions: (a) integrate many previously identified characteristic dimensions and is thus the most inclusive taxonomy to date, (b) were compatible in content with major dimensions of personality traits, and (c) predicted self-reported retrospective behavior (above and beyond personality traits). Although the DIAMONDS still need to be replicated with different data sources, other research has already demonstrated the usefulness of the taxonomy in understanding different personality and person–situation transaction processes. This includes (a) predicting real-time emotion and behavior enacted in situ (Sherman, Rauthmann, Brown, Serfass, & Jones, 2015); (b) unraveling temporal contiguities between personality states and situation experiences in ambulatory assessment data (Rauthmann, Jones, & Sherman, in press); (c) personality traits predicting coming in contact with and uniquely construing situation characteristics (Rauthmann, Sherman, Nave, & Funder, 2015); (d) examining the change in situations within and between individuals (Rauthmann & Sherman, 2016); (e) examining how situational information is dispersed in online social networks, such as Twitter (Serfass & Sherman, 2015); and (f) conceiving situation perception in terms of evolutionarily derived motivational content and processes (Rauthmann, 2016). Further, several validated DIAMONDS measures (32 items: Rauthmann et al., 2014; 24 items: Rauthmann & Sherman, in press[a]; and 8 items: Rauthmann & Sherman, in press[b]) enable a succinct quantification of situations’ major characteristics. Together, it stands to reason that the DIAMONDS can also be usefully applied to a life span perspective on situations. Specifically, we are interested in the age differences in mean levels of the DIAMONDS across the life span.

Mean-Level Patterns of the Situational Eight DIAMONDS

Although there is to date no direct research examining age-graded, mean-level patterns of situation characteristics, research on situation cues and personality development may provide indirect suggestions. However, these indirect lines of research provide competing hypotheses, making it all the more compelling and necessary to empirically examine mean-level patterns across the life span.

Situation cues

Using experience sampling methodology, Wrzus, Wagner, and Riediger (2016) examined, among other things, in a representative German sample of people aged 14–89 years, how often situations occurred in their daily lives containing situation cues related to other persons (alone, family, friends, colleagues, and stranger) and activities (work, chores, leisure, TV/nothing, and conversation). Because we know which cues coincide with which DIAMONDS (see Rauthmann et al., 2014, Table 5), we can derive hypotheses regarding age patterns of the DIAMONDS. First, although situations with family seemed to increase (especially after 60 years), most social cues decreased with age and “alone situations” increased from 50 years onward. This suggests that Sociality may overall decrease across the life span. Second, situations with work decreased and “pastime situations” increased somewhat (e.g., conversations, TV, and leisure). This suggests that, across the life span, Duty and Negativity may overall decrease and pOsitivity increase. Notably, the relations these data suggest pertain to a German sample, so we would expect them to hold at least in Germany. However, only limited inferences of cues to situation characteristics are possible, and we could not generate hypotheses for all DIAMONDS. As such, we turn to the richer literature on personality development to derive further hypotheses.

Personality development

Persons can regulate and affect their own development (e.g., Haase, Heckhausen, & Wrosch, 2013) by, for example, seeking, modifying, and creating situations and environments that are congruent with their personalities (Ickes, Snyder, & Garcia, 1997; Rauthmann & Sherman, 2016; Scarr & McCartney, 1983). These tendencies, in turn, may lead to increased person–environment fit (Harms, Roberts, & Winter, 2006; Rauthmann, 2013; Roberts & Robins, 2004), where certain personality traits are coupled with certain situation characteristics (Rauthmann et al., 2015). For example, the corresponsive principle of personality development (Roberts, Caspi, & Moffitt, 2003) states that traits engender certain situations, which then serve to deepen these very traits. As Rauthmann et al. (2014), Rauthmann et al. (2015), Sherman, Rauthmann, Brown, Serfass, and Jones (2015), and Rauthmann, Jones, and Sherman (in press) have conceptually and empirically demonstrated, the DIAMONDS situation characteristics are contentwise commensurate with the Big Five personality traits, capturing common or similar content: Duty ≅ Conscientiousness, Intellect ≅ Openness, Adversity ≅ Disagreeableness, Mating ≅ Extraversion, pOsitivity ≅ Extraversion and Emotional Stability, Negativity ≅ Neuroticism and Introversion, Deception ≅ Disagreeableness, and Sociality ≅ Agreeableness and Extraversion.

Based on these couplings of personality trait and situation characteristic content and the idea that situations become “synchronized” to one’s personality and vice versa, we could expect the DIAMONDS to follow similar mean-level patterns across the life span to their Big Five counterparts. We thus surveyed major studies on mean-level personality stability (Donnellan & Lucas, 2008; Roberts, Walton, & Viechtbauer, 2006; Specht, Egloff, & Schmulke, 2011; Srivastava, John, Gosling, & Potter, 2003) and summarized derived hypotheses in Table 1. The table is organized around the Big Five and the different DIAMONDS representing their content. As can be seen, we would expect Intellect, Adversity, Mating, and Deception to tendentially decline, while Duty and Sociality to increase. Because findings on the mean-level patterns of Neuroticism have been mixed, Negativity and pOsitivity could either increase or decrease. Note that these hypotheses make different predictions than those derived from situation cues, where Duty and Sociality were expected to decline, not increase.

Table 1.

Summary of Hypotheses and Findings.

Situational Eight	Hypotheses		Patterns Found
Situational Eight	Wrzus et al. (2016): German Situation Cues	Personality Development Literature: International Average Patterns	U.S. Samples	German sample	Similar ?^a
Duty	Decrease	Increase (C)	Increase + decrease	Increase + decrease	Y
Intellect	?	Decrease (O)	Decrease	Decrease	Y
Adversity	?	Increase (E: SD)	Increase + decrease	No change	N
Adversity	?	Decrease (−A)	Increase + decrease	No change	N
Mating	?	Decrease (E: S)	Decrease + increase	Decrease + increase	Y
pOsitivity	Increase	Decrease (E: S)	Decrease + increase	Decrease + increase	N
pOsitivity	Increase	Increase (−N)	Decrease + increase	Decrease + increase	N
Negativity	Decrease	Increase (−E: S; N)	Increase + decrease	Increase + decrease	Y
Negativity	Decrease	Decrease (N)	Increase + decrease	Increase + decrease	Y
Deception	?	Decrease (−A)	Increase + decrease	Decrease	N
Sociality	Decrease	Increase (E: S; A)	Increase + decrease	Decrease	N

Note. O = Openness/Intellect; C = Conscientiousness; E: S = Extraversion: Sociability, E: SD = Extraversion: Social Dominance; A = Agreeableness; N = Neuroticism. A minus (−) signifies that the domain has to be reverted. A question mark (?) signifies that no similar pattern was found.

^a Evaluated based on Figure 1 with an upper limit of 85 years.

Present Research

The present research tested hypotheses generated from extant literature on situation cues and personality development (Table 1). Specifically, we explored the relationship between situation characteristics and age using archival data from annual surveys from two different countries, the American Time Use Survey (ATUS) and the German Socioeconomic Panel (GSOEP). These surveys have four distinct qualities that make them exceptionally suited for our research. First, they have relatively large samples sizes, making it possible to estimate effect sizes precisely. Second, they cover a wide age range, making it possible to study patterns across almost the entire life span from adolescence to old age (15–95 years). Third, the samples are also representative, making it possible to draw generalizations for each country.

Finally, the surveys did not directly sample individuals’ experiences of situations¹ but instead tapped their daily activities (i.e., what participants were doing at a particular hour of the day). Activities—which describe what was going on and what people did—constitute central situation cues and thus are strongly indicative of people’s situations (Rauthmann et al., 2015a; Saucier, Bel-Bahar, & Fernandez, 2007; Sherman, Nave, & Funder, 2010). To obtain situation characteristics data, we had external and trained coders rate all activities on how much they reflected DIAMONDS characteristics. This procedure alleviates the need for in situ ratings by participants directly in the situation, which may otherwise be contaminated by a number of distortion processes. For example, using participants’ own in situ ratings would be problematic because any patterns across the life span could mean changes in response style, biases, memory, or any other factor besides substantive change in characteristics. Further, as Rauthmann et al. (2015a, 2015b) made clear, sampling situation characteristics for each person within their own situation is problematic because one cannot separate situation contact from construal (see Rauthmann, Sherman, et al., 2015). As a result, it is prudent to start with ex situ data rated by external coders as done here. This ensures that activities are coded in similar (i.e., regardless of the age of participants) and reliable ways to reflect what ordinarily socially competent people would think (providing consensual and not idiosyncratic situation data). We thus examined mean levels of actual, third-party–rated situation characteristics, as conveyed per activities, across age in two studies (Study 1: United States and Study 2: Germany).

Study 1

The ATUS is a yearly survey conducted by the U.S. Census Bureau designed to understand how Americans spend their time (additional information can be found at http://www.bls.gov/tus/). Households are randomly selected to participate and are intended to comprise a representative sample of the U.S. population. For each household, an individual who is at least 15 years of age or older is selected to participate. For the present study, we analyzed data from the 2013 ATUS (Sample A) and the most recent 2014 ATUS (Sample B) in an attempt to replicate findings.

Method

Participants

Data from 11,384² individuals who participated in the 2013 ATUS served as the participants for Sample A. The gender breakdown is 55% female with a mean age of 48.28 years (SD = 17.91). The ethnic distribution of the sample is 1% American Indian/Alaskan Native, 4% Asian, 15% Black, 1% Hawaiian/Pacific Islander, 79% White, and <1% other/mixed race.

Data for Sample B come from 11,592 participants in the 2014 ATUS, the most recent year for which data were available. The gender breakdown is 56% female with a mean age of 48.80 years (SD = 17.89). The ethnic distribution of the sample is <1% American Indian/Alaskan Native, 4% Asian, 15% Black, <1% Hawaiian/Pacific Islander, 79% White, and 1% other/mixed race.

Procedure

Participants were contacted by phone and asked to recall all of their daily activities using a Day Reconstruction Method (e.g., Kahneman, Krueger, Schkade, Schwarz, & Stone, 2004). Specifically, interviewers asked participants to report their activities for the preceding 24-hr period with the following prompt: “Yesterday, at 4:00 a.m., what were you doing? What did you do next?” Weekdays were sampled at a rate of approximately 10% each day, with Saturday and Sunday sampled 25% each. Participants reported on average 17.95 and 18.29 activities³ throughout their day, for Samples A and B, respectively (SD _{Sample A} = 7.70; SD _{Sample B} = 7.79). Following this interview, trained coders at the U.S. Census Bureau categorized the participants’ activities according to a three-tiered hierarchy of activity codes. For example, the highest tier—major categories—is comprised of 17 basic activities, such as “personal care” or “socializing, relaxing, and leisure.” Because the lowest tier, encompassing 465 activity codes, afforded the greatest level of specificity (e.g., washing, dressing, and grooming oneself and attending or hosting parties/receptions/ceremonies), we chose to analyze the activities at this fine-grained level.

Measures and Coding Strategy

Participants self-reported their daily activities but did not actually rate their situations. Thus, it was necessary to create prototypical DIAMONDS profiles for each of the activity codes. For example, if a participant’s activity was coded as “relaxing and thinking,” we needed to determine to what extent different characteristics would be relevant or salient in a situation with that activity. We excluded all activity codes that indicated the participant was sleeping and those that could not be coded in any meaningful way (e.g., catchall codes that indicate “not elsewhere classified”), leaving 351 psychologically meaningful activity codes to be rated. Four trained research assistants rated all 351 activity codes using the 24-item S8* (Rauthmann & Sherman, in press[a]). Coders indicated whether each item was relevant to the activity code on a 5-point Likert-type scale from 1 (strongly disagree) to 5 (strongly agree).

We calculated interrater reliability using intraclass correlation coefficients (ICCs; Shrout & Fleiss, 1979) to determine to what extent the coders generally agreed on the prototypical situation profile for each activity code.⁴ The average item-level agreement ICC(3,k) across the 24 items was .51 (SD = .26), which resulted in an average reliability of the composites of .57 (SD = .19). Because the ICCs indicated good agreement, we then created DIAMONDS composite profiles by averaging across the four raters for each respective activity code. Next, we converted each participant’s activities to the respective composite profile. To aid in the interpretability of the results due to the arbitrary nature of the coding scale, we converted all DIAMONDS scores to percentage of maximum possible (POMP) scores (Cohen, Cohen, Aiken, & West, 1999). Finally, we created a single profile for each age-group by aggregating over all situations by age.⁵

Descriptives (M, SD, and α) and intercorrelations (between-person and within-person) are presented in Table 2. Aggregated over all situations, there were varying base rates for each of the DIAMONDS. For example, most situations were characterized by more Duty (M = 66.05) but by less Adversity (M = 2.10) and Deception (M = 4.92). These rates are consistent with a recent experience-sampling study employing the DIAMONDS (Sherman et al., 2015).

Table 2.

Descriptive Statistics and Intercorrelations.

Dimensions	Descriptives			Intercorrelations
Dimensions	M	SD	α^a	D	I	A	M	O	N	D	S
Study 1: ATUS 2013
Duty	66.05	8.49	.88	—	.26	.44	.21	−.04	.41	−.07	.16
Intellect	36.84	8.11	.97	.54	—	.37	.05	.18	.65	.27	.71
Adversity	2.10	1.00	.61	.49	.28	—	.27	.03	.52	.27	.21
Mating	8.38	2.76	.50	.07	−.13	.17	—	.47	−.15	−.27	−.03
pOsitivity	37.07	6.95	.93	−.13	−.09	−.06	.48	—	−.19	−.06	.28
Negativity	19.91	7.32	.88	.63	.76	.45	−.28	−.33	—	.43	.56
Deception	4.92	1.92	.82	−.01	.26	.22	−.28	−.19	.37	—	.23
Sociality	35.29	10.38	.86	.43	.69	.08	.11	.23	.50	.06	—
Study 1: ATUS 2014
Duty	66.13	8.49	.88	—	.26	.44	.21	−.05	.41	−.07	.16
Intellect	36.88	8.14	.97	.54	—	.37	.06	.18	.65	.27	.71
Adversity	2.12	0.99	.61	.50	.29	—	.27	.03	.53	.27	.21
Mating	8.43	2.74	.50	.07	−.13	.17	—	.47	−.14	−.27	−.03
pOsitivity	37.08	6.90	.93	−.14	−.07	−.05	.48	—	−.20	−.07	.28
Negativity	20.06	7.32	.88	.63	.75	.44	−.28	−.32	—	.43	.55
Deception	4.95	1.93	.82	−.03	.26	.22	−.29	−.18	.37	—	.23
Sociality	35.18	10.41	.86	.42	.69	.08	.12	.24	.49	.05	—
Study 2: SOEP-IS
Duty	50.02	13.24	.98	—	.19	.11	.02	−.57	.80	−.08	.08
Intellect	17.79	8.07	.84	.33	—	.46	−.27	−.16	.31	.39	.05
Adversity	0.89	1.14	.80	.09	.28	—	.46	−.27	−.16	.31	.39
Mating	11.36	3.62	.87	−.07	.09	.07	—	.22	.00	.38	.16
pOsitivity	42.69	6.58	.84	−.72	−.45	−.21	.18	—	−.74	−.09	.26
Negativity	19.23	8.23	.98	.84	.40	.30	.07	−.80	—	.28	.13
Deception	5.06	4.42	.99	−.09	.44	.29	.72	.04	.18	—	.45
Sociality	22.34	7.16	.95	.18	.11	.16	.42	.03	.36	.55	—

Note. ATUS (2013)—Study 1 (United States), Sample A: N = 11,384. ATUS (2014)—Study 1 (United States), Sample B: N = 11,592. SOEP-IS—Study 2 (Germany): N = 2,299. Means and SDs are reported as POMP scores. Between-person correlations are below the diagonal and weighted within-person correlations are above the diagonal. ATUS = American Time Use Survey; SOEP-IS = Socioeconomic Panel-Innovation Sample; SD = standard deviation; POMP = percentage of maximum possible; D = Duty; I = Intellect; A = Adversity, M = Mating; O = pOsitivity; N = Negativity; D = Deception; S = Sociality.

^aStudy 1: N _raters = 4 and Study 2: N _raters = 2.

Results

To investigate the relationship between age and situation characteristics, we estimated a series of regressions for each DIAMONDS. We first started with a simple linear model and subsequently added quadratic and cubic terms. Consistent with research on the association between personality traits and age (e.g., Donnellan & Lucas, 2008; Srivastava et al., 2003), we only retained curvilinear models if model fit was improved over more parsimonious models by at least F ≥ 25.00. Models were run for the combined sample as well as for each gender separately. Because of largely negligible gender-specific effects, we only report findings from the combined sample.⁶ Regression results appear in Table 3 and are visually displayed in Figure 1 (U.S. Sample A and U.S. Sample B). We first review the results from Sample A. Duty was best characterized by a cubic effect: Younger and older people’s situations showed lower levels of duty, with the highest level in people’s late 30s to early 40s. Intellect also demonstrated a cubic pattern of age differences: Younger people’s situations showed lower levels of intellect, peaking with 40-year-olds before exhibiting a downward linear pattern. Adversity exhibited a cubic effect: Younger people’s situations generally showed lower levels of adversity, which steadily increased until reaching a peak around age 30. Older people’s situations tended to continue to show relatively high levels of Adversity (although there was substantial variability). Mating was best characterized by a positive quadratic effect: 20-year-olds’ situations showed relatively high peak levels while declining thereafter until reaching a trough around age 50. Mating then began to increase again with levels of those in old age similar to people in their 30s. Although pOsitivity appeared to visually demonstrate a cubic pattern, our models indicated that a quadratic model fit best: People’s situations in their 20s tended to show the highest levels of pOsitivity, with the lowest levels for 50 year olds. Further, pOsitivity increased among older people, although it never reached the same level exhibited by 20- to 30-year-olds. Negativity followed a cubic pattern: Younger people’s situations showed minimal levels which, however, tended to increase until the mid-30s and decrease again steadily thereafter. Deception also exhibited a cubic pattern: Younger people’s situations showed low Deception with older people’s situations showing more. Further, this pattern decreased among individuals in their mid-60s. Finally, Sociality showed a cubic pattern: It was higher in younger people with a peak around age 40 before exhibiting a decrease thereafter.

Table 3.

Model Parameters of DIAMONDS Predictions From Age Variables.

Parameters	U.S. Samples		German Sample
Parameters	Sample A (2013)	Sample B (2014)	GSOEP-IS (2012)
Duty
Constant	68.78 (.14)	68.97 (.15)	52.37 (.43)
Age	−0.22 (.01)	−0.23 (.01)	−0.49 (.03)
Age²	−0.005 (.00)	−0.006 (.00)	−0.005 (.00)
Age³	0.0002 (.00)	0.0002 (.00)	0.0003 (.00)
R	.97	.96	.91
F for linear	35.46	29.57	51.19
ΔF for quadratic	372.85	291.23	63.03
ΔF for cubic	182.97	159.99	126.79
Intellect
Constant	38.47 (.11)	38.46 (.11)	17.10 (.22)
Age	−0.16 (.01)	−0.17 (.01)	−0.12 (.01)
Age²	−0.003 (.00)	−0.003 (.00)	—
Age³	0.00008 (.00)	0.00009 (.00)	—
R	.97	.97	.82
F for linear	206.20	219.10	156.10
ΔF for quadratic	162.70	137.17	3.11
ΔF for cubic	51.01	56.84	0.43
Adversity
Constant	2.24 (.01)	2.25 (.01)	0.87 (.03)
Age	−0.005 (.00)	−0.007 (.00)	−0.001 (.00)
Age²	−0.0003 (.00)	−0.0003 (.00)	—
Age³	0.00001	0.00001 (.00)	—
R	.88	.79	.12
F for linear	8.60	4.67	3.42
ΔF for quadratic	122.38	53.52	0.79
ΔF for cubic	59.49	40.73	0.06
Mating
Constant	8.30 (.04)	8.36 (.04)	10.71 (.17)
Age	−0.006 (.00)	−0.008 (.00)	−0.03 (.00)
Age²	0.0005 (.00)	0.0005 (.00)	0.001 (.00)
Age³	—	—	—
R	.68	.73	.68
F for linear	11.85	23.64	31.74
ΔF for quadratic	37.63	40.59	26.41
ΔF for cubic	0.67	9.78	7.74
pOsitivity
Constant	36.35 (.20)	36.47 (.20)	42.48 (.27)
Age	−0.04 (.00)	−0.06 (.01)	0.25 (.02)
Age²	0.002 (.00)	0.002 (.00)	0.001 (.00)
Age³	—	—	−0.0002 (.00)
R	.70	.78	.87
F for linear	22.97	49.49	80.64
ΔF for quadratic	30.68	36.40	7.91
ΔF for cubic	5.36	22.66	65.02
Negativity
Constant	22.22 (.10)	22.36 (.12)	20.00 (.27)
Age	−0.17 (.01)	−0.18 (.01)	−0.36 (.02)
Age²	−0.005 (.00)	−0.005 (.00)	−0.002 (.00)
Age³	0.0001 (.00)	0.0001 (.00)	0.0003 (.00)
R	.98	.97	.93
F for linear	43.45	40.00	77.10
ΔF for quadratic	704.53	419.16	17.52
ΔF for cubic	159.34	116.10	166.24
Deception
Constant	5.06 (.04)	5.02 (.04)	4.85 (.14)
Age	0.03 (.00)	0.03 (.00)	−0.07 (.01)
Age²	−0.0002 (.00)	−0.00002 (.00)	—
Age³	−0.00002 (.00)	−0.00003 (.00)	—
R	.83	.83	.78
F for linear	53.92	58.86	116.00
ΔF for quadratic	8.44	0.69	9.11
ΔF for cubic	39.93	40.59	1.78
Sociality
Constant	37.91 (.17)	37.78 (.17)	22.22 (.28)
Age	−0.31 (.01)	−0.32 (.01)	−0.12 (.01)
Age²	−0.004 (.00)	−0.003 (.00)	—
Age³	0.0002 (.00)	0.0002 (.00)	—
R	.98	.98	.73
F for linear	283.90	331.40	87.50
ΔF for quadratic	92.91	95.08	2.86
ΔF for cubic	89.01	92.19	7.57

Note. Ns are 11,384, 11,592, and 2,299 for U.S. Sample A, U.S. Sample B, and German sample, respectively. Standard errors are indicated in parentheses. Coefficients are provided in unstandardized units. Age was mean-centered before creating curvilinear terms (U.S. samples: M_age = 48.06 and German sample: M_age = 51.80). Model parameters are reported for the final retained model. More complex models were only retained if ΔF was ≥25.00 over a simpler model. Confidence intervals are not presented here because the large sample sizes afforded high precision (e.g., for many effects, they would be reported to the thousandth place). GSOEP-IS = German Socioeconomic Panel-Innovation Sample.

Figure 1.

DIAMONDS situation characteristics across the life span. Note. y axis = DIAMONDS score in percentage of maximum possible (POMP) units and x axis = age. Each point represents the mean DIAMONDS score (in POMP units) aggregated by each age, respectively. Lines represent the slope of the retained model with gray bands indicating the 95% confidence interval. Due to differences in the range of POMP scores between the American and the German samples, the German sample plots are scaled differently. The dashed horizontal line in the German sample plots indicate the upper age limit of the American samples (age 85). When interpreting the confidence bands, it should be noted that the American Time Use Survey did not sample any individuals between 81 and 84 years of age; the German data did not capture 92-year-olds.

We next examined the results from Sample B to determine the robustness of findings from Sample A. As the middle column of Table 2 indicates, all of the patterns in Sample A were replicated in Sample B. That is, participants in 2013 and 2014, on average, showed similar mean-level DIAMONDS patterns across ages.

Discussion

Study 1 provided replicable evidence that situation characteristics are age graded. Overall, however, the patterns provided only mixed support for our a priori hypotheses (left columns in Table 1). The data generally supported our expected patterns for Intellect, pOsitivity, and Negativity; there was only partial support for Adversity and Mating; and Duty and Deception exhibited patterns we had not previously expected (summarized in the right-side columns of Table 1). Because it is possible that these findings may be limited by everyday activities that occur only within the United States and thus reflect country-specific patterns, we sought to replicate them in a representative German sample in Study 2.

Study 2

Study 1 demonstrated a robust relationship between major situation characteristics and age in two representative samples from the United States. But are the daily activities and situations by U.S. Americans similar to those of people from other countries? Recent findings from cross-cultural situation research would suggest that situations are rather similar across the world (Guillaume et al., in press). To investigate whether age differences in situation characteristics are indeed generalizable to a non-U.S. sample, we analyzed data from a large representative sample of Germans that also included a wider age range (18–95 years compared to 15–85 years in the U.S. samples).

Participants and Procedure

Data for Study 2 come from 2,299 individuals (M _age = 51.8 years, SD = 18; 48% male) who participated in the 2012 GSOEP-Innovation Sample (IS). That sample is a subset of the GSOEP, a yearly economic survey of German households. Participants engaged in a Day Reconstruction Method task by first segmenting their prior day into a series of episodes and indicating when an activity started and ended (Anusic, Lucas, & Donnellan, in press). Each activity was already coded within the GSOEP-IS into 1 of the 22 activity categories, which included “shopping,” “preparing food,” and “working/studying.”⁷ We excluded two activity codes because they could not be meaningfully coded (“end of the day: beginning of night rest” and “other”). Participants reported on average 9.98 activities (SD = 3.84). Further details regarding the Day Reconstruction Method in the GSOEP-IS can be found in Anusic, Lucas, and Donnellan (in press).

Measures and Coding Strategy

None of the activities was rated in terms of situation characteristics. Thus, similar to Study 1, it was necessary to create prototypical DIAMONDS profiles for each of the GSOEP-IS activity codes. The authors rated each activity code on the S8* using a 5-point Likert-type scale (see Measures and Coding Strategy section in Study 1). The average item-level agreement ICC(3,k) across the 24 items was .70 (SD = .21), which resulted in an average reliability of the composites of .75 (SD = .16; see note 4). As with Study 1, participants’ activities were converted to a composite profile, where DIAMONDS scores were transformed to POMP scores. Finally, we aggregated across all profiles by age.

Results and Discussion

As in Study 1, we estimated a series of linear regressions predicting each DIAMONDS from age. We again estimated linear, quadratic, and cubic models and only retained a more complex model if the F value was ≥25.00 over a simpler model. The parameters from the regressions appear in the far right-hand column of Table 3 and are visually displayed in the far right panel of Figure 1.

Overall, four of the DIAMONDS—Duty, Intellect, Mating, and Negativity—exhibited a similar pattern to those found in the U.S. data. For example, Duty demonstrated a cubic effect, such that younger people’s situations showed initially lower levels, peaking among those in their 40s and decreasing thereafter. Negativity similarly demonstrated a cubic pattern such that it increased among people in their 20s through 40s, ultimately peaking among 40-year-olds. After decreasing thereafter, Negativity then exhibited an increase among the elderly.

The patterns for Adversity, pOsitivity, Deception, and Sociality did not replicate in the German sample. For example, no pattern was observed for Adversity as indicated by the flat slope in Figure 1. The pOsitivity showed a cubic pattern in the German sample which was quite different from the U.S. pattern: It was initially high among 20-year-olds, decreased steadily thereafter, reached a low among 40-year-olds, and steadily increased thereafter again. Deception and Sociality were both characterized by negative linear patterns such that younger people’s situations showed the highest levels before subsequently decreasing in older people.

General Discussion

This research aimed to examine mean-level patterns of major situation characteristics, the Situational Eight DIAMONDS, over the life span. Across three samples and two countries, we found evidence that situation characteristics were indeed age graded, that is, individuals tended to be in different kinds of situations as a function of their age. Table 1 succinctly summarizes predictions and actual findings from the two U.S. samples and the German sample.

Interpretation

Prior research has demonstrated that certain stages of development with their respective tasks (e.g., finding a romantic partner) and normative (nonrandom) life events (e.g., marriage) co-occur in a given socioculture in an age-graded manner (Bleidorn et al., 2014). Developmental tasks and especially life events may indeed be seen as special or intense situational episodes and can thus be also described with situation characteristics (Rauthmann et al., 2015b). Furthermore, according to the social investment model, individuals also assume new roles—such as a spouse, parent, employee, and grandparent—at times that are age graded (Lodi-Smith & Roberts, 2007). These role transitions might also afford opportunities for situations that were not possible before (e.g., taking care of a child, reporting to work every day, etc.). In the current research, these differences in situations might also be driven by parallel age differences in activities. Follow-up analyses demonstrated that there are in fact sensible age differences at the activities level. For example, in Study 2, middle-aged adults tended to report caring for children more often relative to older participants. Similarly, adults in their 70s and 80s reported spending more time watching TV compared to young-aged and middle-aged adults. Thus, situation characteristics may vary as a function of age because of the range of situations and activities that one can experience on a daily basis in particular roles (Jones, Brown, Serfass, & Sherman, 2014). At what age and how such roles are assumed are, in turn, governed by sociocultural norms, standards, and values that can partially drive observed differences in mean-level patterns between different countries.

Drawing on Wrzus and colleagues (2016) and the extant personality development literature (e.g., Roberts et al., 2006), we tested a number of predictions regarding situation characteristics and age (Table 1). For example, because prior research showed that Openness tends to decrease across the life span (Roberts et al., 2006; Srivastava et al., 2003), we expected a likewise decrease in Intellect in older people—which is what we found in all samples. On the other hand, based on mean-level change in Conscientiousness (e.g., increasing as people age), we expected to observe a similar increase in Duty—which we found for neither sample. However, when not considering the personality development predictions but instead the situation cues predictions derived from Wrzus and colleagues’ (2016) findings, the German data did somewhat show the expected patterns of decreasing Duty, Negativity, and Sociality and increasing pOsitivity (Table 1, third panel in Figure 1). This may not be surprising as both this and Wrzus and colleagues’ study used a representative German sample. Indeed, this may underscore that there are differences in mean-level patterns of situation characteristics across countries.

Limitations and Future Research

This is the first study to examine age differences patterns in situation characteristics. Although the large data sets analyzed provide unique insights into how people spend their time, they are necessarily limited. First, situations were coded using predetermined activity categories which we used as proxies for situational information. It was not possible to have access to or rate the actual situations that participants experienced. More comprehensive situation descriptions (tapping into a broader range of situation cues, not just activities) could have yielded more situational information for the coders to rate.

Second, the activity categories differed in number and breadth across the two studies, which may limit the comparability of the patterns. However, it should be noted that the 351 categories of Study 1 are at the lowest tier and are comprised of 17 “major categories” at the highest level. Indeed, these “major categories” (Study 1) overlap substantially with those used in Study 2 (see Supplemental Table 1 online).

Third, by additionally gathering in situ ratings from participants themselves (i.e., ratings from people directly in the situation), we would be able to test whether the same age patterns emerge as in Figure 1. Future research should thus try to replicate the patterns found for people’s situation experiences.

Fourth, using the prototypical DIAMONDS profiles for every person with the same activity codes can be problematic because interindividual differences are washed out. By doing so, it is not possible to examine within-country variation for the “same” activity/situation. Additionally, more countries should be examined to sensibly examine cross-cultural similarities or differences (Guillaume et al., in press).

Fifth, overall, the effect sizes for the observed linear patterns were modest in absolute terms, ranging from |.005| (Adversity, U.S. Sample A) to |.49| (Duty, German sample). The curvilinear terms for the effects were substantially smaller, but this is not uncommon in this type of research (e.g., Donnellan & Lucas, 2008). However, the large sample sizes of these data yielded extremely narrow confidence intervals and relatively precise effect size estimates.

Finally, the data presented here are purely cross-sectional in nature and thus it is not possible to examine developmental trajectories within people or to rule out alternative explanations such as differences between cohorts. To that end, interpretations regarding the mean-level patterns over age are speculative and should be treated as such. Future studies should use longitudinal designs with different cohorts to answer developmental questions of stability and change in people’s situations.

Taken together, the limitations of this work provide suggestions for future research. Specifically, future examinations of age patterns of situation characteristics should (a) employ better situation descriptions (ideally not just verbally but also with pictures or video clips), (b) make use of both situation ratings performed by participants in the situations and by external raters, (c) sample more and different countries, (d) examine variance across countries (intercultural differences) and within countries (interindividual differences), and (e) make use of longitudinal designs.

Conclusion

This research demonstrates that situation characteristics are age graded such that there are age differences in mean-level patterns across the life span. It is plausible that these patterns are related to the opportunities and constraints for situations that emerge at various ages across the life span, afforded by roles that are governed by sociocultural rules, norms, and values. As such, the situations that people encounter in their everyday lives may partially be driven by changing roles, developmental tasks, and life events that are culturally influenced. As a result, major dimensions of situation characteristics, such as the Situational Eight DIAMONDS, evince each specific mean-level patterns across the life span.

Footnotes

Acknowledgments

We thank M. Brent Donnellan for his helpful advice regarding statistical analyses. We also thank Ryne Sherman for providing resources for coding the American Time Use Survey data and his helpful comments on a previous version of this manuscript. Any errors or omissions remain our own. All analyses were conducted in R (R Core Team, 2015).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Supplemental Material

The online data supplements are being hosted by the Open Science Framework and are available at .

References

Anusic

Lucas

R. E.

Donnellan

M. B.

(in press). The validity of the Day Reconstruction Method in the German Socio-economic Panel Study. Social Indicators Research.

Bleidorn

Klimstra

T. A.

Denissen

J. J. A.

Rentfrow

P. J.

Potter

Gosling

S. D.

(2014). Personality maturation around the world: A cross-cultural examination of Social Investment Theory. Psychological Science, 24, 2530–2540.

Cohen

Aiken

L. S.

West

S. G.

(1999). The problem of units and the circumstance for POMP. Multivariate Behavioral Research, 34, 315–346.

de Raad

(2004). Situations that matter to personality. In Eliasz

Hampson

S. E.

de Raad

(Eds.), Advances in personality psychology (Vol. 2, pp. 179–204). Hove, England: Routledge.

Donnellan

M. B.

Lucas

R. E.

(2008). Age differences in the Big Five across the lifespan: Evidence from two national samples. Psychology and Aging, 23, 558–566.

Edwards

J. A.

Templeton

(2005). The structure of perceived qualities of situations. European Journal of Social Psychology, 35, 705–723.

Guillaume

Baranski

Bastian

Bronin

Ivanova

Cheng

J. T.

… Funder

D. C.

(in press). The World at 7: Comparing the experience of situations across 20 countries. Journal of Personality.

Haase

Heckhausen

Wrosch

(2013). Developmental regulation across the life span: Toward a new synthesis. Developmental Psychology, 49, 964–972.

Harms

P. D.

Roberts

B. W.

Winter

(2006). Becoming the Harvard man: Person-environment fit, personality development, and academic success. Personality and Social Psychology Bulletin, 32, 851–865.

10.

Ickes

Snyder

Garcia

(1997). Personality influences on the choice of situations. In Hogan

Johnson

Briggs

(Eds.), Handbook of personality psychology (pp. 165–195). San Diego, CA: Academic Press.

11.

Jones

A. B.

Brown

N. A.

Serfass

D. G.

Sherman

R. A.

(2014). Traumatic experiences yield changes in the situation experience. European Journal of Personality, 28, 339–340. doi:10.1002/per.1970

12.

Kahneman

Krueger

A. B.

Schkade

D. A.

Schwarz

Stone

(2004). A survey method for characterizing daily life experience: The Day Reconstruction Method. Science, 306, 1776–1780.

13.

Lodi-Smith

Roberts

B. W.

(2007). Social investment and personality: A meta-analysis of the relationship of personality traits to investment in work, family, religion, and volunteerism. Personality and Social Psychology Review, 11, 68–86.

14.

R Core Team. (2015). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.

15.

Rauthmann

J. F.

(2013). Effects of supplementary and complementary personality-situation fit on personality processes. Psychology of Everyday Activity, 6, 41–63.

16.

Rauthmann

J. F.

(2016). Motivational factors in the perception of psychological situation characteristics. Social and Personality Psychology Compass, 10, 92–108. doi: 10.1111/spc3.12239

17.

Rauthmann

J. F.

Gallardo-Pujol

Guillaume

E. M.

Todd

Nave

C. S.

Sherman

R. A.

… Funder

D. C.

(2014). The Situational Eight DIAMONDS: A taxonomy of major dimensions of situation characteristics. Journal of Personality and Social Psychology, 107, 677–718.

18.

Rauthmann

J. F.

Jones

A. B.

Sherman

R. A.

(in press). Directionality of person-situation transactions: Are there spill-overs among and between situation experiences and personality states? Personality and Social Psychology Bulletin.

19.

Rauthmann

J. F.

Sherman

R. A.

(2016). Situation change: Stability and change of situation variables between and within persons. Frontiers in Psychology, 6, 1938.

20.

Rauthmann

J. F.

Sherman

R. A.

(in press[a]). Measuring the Situational Eight DIAMONDS characteristics of situations: An optimization of the RSQ-8 to the S8*. European Journal of Psychological Assessment.

21.

Rauthmann

J. F.

Sherman

R. A.

(in press[b]). Succinctly measuring the psychological characteristics of situations: Ultra-brief measures for the Situational Eight DIAMONDS domains. European Journal of Psychological Assessment.

22.

Rauthmann

J. F.

Sherman

R. A.

Funder

D. C.

(2015a). Principles of situation research: Towards a better understanding of psychological situations. European Journal of Personality, 29, 363–381.

23.

Rauthmann

J. F.

Sherman

R. A.

Funder

D. C.

(2015b). New horizons in research on psychological situations and environments. European Journal of Personality, 29, 382–432.

24.

Rauthmann

J. F.

Sherman

R. A.

Nave

C. S.

Funder

D. C.

(2015). Personality-driven situation experience, contact, and construal: How people’s personality traits predict characteristics of their situations in daily life. Journal of Research in Personality, 55, 98–111.

25.

Reis

H. T.

(2008). Reinvigorating the concept of situation in social psychology. Personality and Social Psychology Review, 12, 311–329.

26.

Roberts

B. W.

Caspi Moffit

(2003). Work experiences and personality development in young adulthood. Journal of Personality and Social Psychology, 84, 582–593.

27.

Roberts

B. W.

Pomerantz

E. M.

(2004). On traits, situations, and their integration: A developmental perspective. Personality and Social Psychology Review, 8, 402–416.

28.

Roberts

B. W.

Robins

R. W.

(2004). Person-environment fit and its implications for personality development: A longitudinal study. Journal of Personality, 72, 89–110.

29.

Roberts

B. W.

Walton

K. E.

Viechtbauer

(2006). Patterns of mean-level change in personality traits across the life course: A meta-analysis of longitudinal studies. Psychological Bulletin, 132, 1–25.

30.

Saucier

Bel-Bahar

Fernandez

(2007). What modifies the expression of personality tendencies? Defining basic domains of situation variables. Journal of Personality, 75, 479–504.

31.

Scarr

McCartney

(1983). How people make their own environments: A theory of genotype → environment effects. Child Development, 54, 424–435.

32.

Serfass

D. S.

Sherman

R. A.

(2015). Situations in 140 characters: Assessing real-world situations on Twitter. PLOS One, 10, 1–19.

33.

Sherman

R. A.

Nave

C. S.

Funder

D. C.

(2010). Situational similarity and personality predict behavioral consistency. Journal of Personality and Social Psychology, 99, 330–343.

34.

Sherman

R. A.

Rauthmann

J. F.

Brown

N. A.

Serfass

D. S.

Jones

A. B.

(2015). The independent effects of personality and situations on real-time expressions of behavior and emotion. Journal of Personality and Social Psychology, 109, 872–888.

35.

Shrout

P. E.

Fleiss

J. L.

(1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86, 420–428.

36.

Specht

Egloff

Schmulke

S. C.

(2011). Stability and change of personality across the life course: The impact of age and major life events on mean-level and rank-order stability of the Big Five. Journal of Personality and Social Psychology, 101, 862–882.

37.

Srivastava

John

O. P.

Gosling

S. D.

Potter

(2003). Development of personality in early and middle adulthood: Set like plaster or persistent change? Journal of Personality and Social Psychology, 84, 1041–1053.

38.

Wrzus

Wagner

G. G.

Riediger

(2016). Personality-situation transactions from adolescence to old age. Journal of Personality and Social Psychology, 110, 782–799.