Items

Table 1

Item Wording of the Risk Proneness Short Scale (R-1)

Response specifications

The item is answered using a 7-point rating scale with labeled endpoints ranging from not at all willing to take risks (1) to very willing to take risks (7).

Scoring

R-1 (Risk Proneness Short Scale) consists of one item covering the psychological disposition of risk proneness. The English-language adaptation of this item is displayed in Table 1 (for the original German item, see Beierlein, Kovaleva, Kemper, et al., 2015). As in the German-language source instrument, the item is positively worded in relation to the underlying construct. The scale score of risk proneness is equivalent to the respondent’s answer to the item.

Application field

The Risk Proneness Short Scale measures risk proneness in an economical and time-efficient way. As an ultra-short scale with a completion time of < 10 s (estimated value), R-1 is applicable in a variety of research areas. It is particularly well suited for research settings in which there are severe time limitations or constraints to questionnaire length. R-1 was originally developed in German (see Beierlein, Kovaleva, Kemper, et al., 2015) and validated in a large and diverse random sample of German adults. In the present study, it was adapted to the English language and validated in the United Kingdom, while paying heed to different age groups, genders, and social classes. R-1 is typically self-administrated, such as via paper-and-pencil or online questionnaire. However, provided slight adaptations to the instructions, an oral administration in a personal interview or telephone interview is also conceivable.

https://doi.org/10.6102/zis324_exzRisk proneness denotes the willingness to take risks, that is, a general preference to choose (or avoid) risky behavioral options (e.g., Brockhaus, 1980; Sitkin & Pablo, 1992). Risk proneness is a rather stable personality trait (Zuckerman et al., 1980; test–retest stability r_tt = .45–.53 across 5 to 10 years, Josef et al., 2016). It affects how individuals evaluate the probability, value, and amount of potential gains and losses of a decision under uncertain conditions. The consequences of a (potential risky) decision are always compared to the status quo or another reference point (Kahneman & Tversky, 1979). Thus, concrete behavior in a risk situation depends both on behavioral predispositions (e.g., risk preferences, outcome history) and on situational factors (e.g., problem framing, social influence; Sitkin & Pablo, 1992). Self-reported risk proneness relates to risky behavior in different areas of life such as finance, career, and health (e.g., Dohmen et al., 2011).

Risk proneness also differs with regard to different sociodemographic groups: Men tend to be more willing to take risks than women (e.g., Dohmen et al., 2011; Harris et al., 2006). An explanation for these gender differences might be that women compared to men consider the probability of negative consequences of risky behavior as higher and perceive possible gains from risky options (e.g., enjoyment) as smaller (Harris et al., 2006). The willingness to take risks and the tendency to actually engage in risky behavior are also related to age: Older people have a lower willingness to take risks than younger people (e.g., Dohmen et al., 2011). Apart from associations with sociodemographic characteristics, generally, high risk proneness is positively correlated with related personality traits and motivational orientations such as sensation seeking (e.g., Breivik et al., 2017; for the concept of sensation seeking, see Zuckerman, 2007), the Big Five domains of Extraversion, Openness, and Emotional Stability (e.g., Tok, 2011), general self-efficacy (e.g., Olivari et al., 2018), internal locus of control (e.g., Lemarié et al., 2019), optimism (e.g., Barel, 2019), life satisfaction (e.g., Muffels & Headey, 2013), and self-rated health (e.g., Dohmen et al., 2011).

Because the willingness to take risks is a good predictor for multiple constructs (see paragraph above), it is regularly included in multiple surveys (e.g., in the German Longitudinal Election Study [GLES] or in the German Socioeconomic Panel [SOEP]). Risk proneness is usually measured in surveys using self-report measures. Although domain-specific risk proneness measures (e.g., willingness to take financial risks) can increase the prediction of domain-specific risk-taking behavior (e.g., investing in highly volatile markets), domain-general risk proneness measures have significant predictive power to explain variation in domain-specific risk-taking behavior (e.g., Dohmen et al., 2011). Therefore, Beierlein, Kovaleva, Kemper, et al. (2015) reanalyzed and revised a German-language single-item scale measuring domain-general risk proneness (i.e., the one usually surveyed in SOEP; Richter et al., 2013) to arrive at a psychometrically sound ultra-short (single-item) scale.

Item generation and selection

To develop the German-language source version of R-1, Beierlein, Kovaleva, Kemper, et al. (2015) drew on an existing single-item measure for measuring risk proneness (Richter et al., 2013) that had been used in the longitudinal German survey SOEP. After the scale was pre-tested cognitively, the item and its rating scale were revised to align them with best practices of scale construction (e.g., by reducing the number of response categories; for more detailed information, see Beierlein, Kovaleva, Kemper, et al., 2015). The German-language R-1 was thoroughly validated based on a large and diverse random sample representing the adult population in Germany in terms of age, gender, and educational attainment.

Because researchers may be interested in comparing the level of risk proneness between different societies, there is a need for a cross-culturally valid measure. To enhance the usability of R-1, and to enable social surveys to use R-1 in an English-language context, the scale was adapted to the English language (by Beierlein, Kovaleva, Kemper, et al., 2015) and validated in a sample from the UK (in the present study). First, the one item of R-1 and its rating scale were adapted to English by translating the item and its rating scale following the TRAPD approach (Translation, Review, Adjudication, Pretesting, and Documentation; Harkness, 2003), whereby two professional translators (English native speakers) translated the item and the rating scale independently of each other into British English and American English, respectively. Second, an adjudication meeting was held in which psychological experts, the two translators, and an expert in questionnaire translation reviewed the translation proposals and developed the final translation.

The source instrument by Beierlein, Kovaleva, Kemper, et al. (2015) was developed in and validated for the German language. The aim of the present study was to validate the English-language adaptation of R-1 and to directly compare its psychometric properties with those of the German-language source version.

Samples

To investigate the psychometric properties of the English-language adaptation of R-1 and their comparability with those of the German-language source instrument, we assessed both versions in a web-based survey (using computer-assisted self-administered interviewing [CASI]) conducted by the online access panel provider respondi AG in the UK and Germany (DE), respectively. Fielding took place in January 2018. For both the UK and Germany, quota samples were drawn that represented the heterogeneity of the adult population with regard to age, gender, and educational attainment. Only native speakers of the respective languages were recruited. We explained our research goal (investigation of the quality of several questionnaires) to the participants. Respondents were financially rewarded for their participation. In both nations, a subsample was reassessed after approximately 3 to 4 weeks (median time intervals: 28 days in the UK and 20 days in Germany).

Only respondents who completed the full questionnaire—that is, who did not abort the survey prematurely—were included in our analyses. To handle missing values on single items (of R-1 or other variables included in the survey), we used full information maximum likelihood estimation (FIML) in our analyses. The gross sample sizes were N_UK = 508 and N_DE = 513. In the next step, invalid cases were excluded based on (a) ipsatized variance—that is, the within-person variance across items (Kemper & Menold, 2014)—if the person fell within the lower 5% of the sample distribution of ipsatized variance; (b) the Mahalanobis distance of a person’s response vector from the average sample response vector (Meade & Craig, 2012) if the person fell within the upper 2.5% of the sample distribution of the Mahalanobis distance; and (c) implausibly short response times, namely, if the person took, on average, less than 1 s to respond to an item. Our intention in choosing relatively liberal cut-off values was to avoid accidentally excluding valid cases. All exclusion criteria were applied simultaneously, that is, any respondent who violated one or more of the three criteria was excluded from the analyses and that only those who met all three criteria were included. This approach resulted in 40 cases (7.9%) being excluded from the UK subsample and 39 cases (7.6%) from the German subsample, yielding net sample sizes of N_UK = 468 (retest: N_UK = 111) and N_DE = 474 (retest: N_DE = 117). Table 2 depicts in detail the sample characteristics and their distribution.

Table 2

Sample Characteristic Features

Note. The equivalent German educational levels were as follows (from low to high): ohne Bildungsabschluss/Hauptschule [no educational qualification; lower secondary leaving certificate]; mittlerer Schulabschluss [intermediate school leaving certificate]; (Fach-)Hochschulreife [higher education entrance qualification].

Material

Online surveys were conducted in German for the German sample and in English for the UK sample. They comprised the respective language version of R-1. In addition, the following short scale measures were also administered as part of the survey to validate R-1 against (a) the Big Five personality traits, (b) impulsive behavior, (c) optimism–pessimism, (d) general self-efficacy, (e) self-esteem, (f) locus of control, (g) socially desirable responding, (h) life satisfaction, and (i) health, respectively:

(a)     the extra-short form of the Big Five Inventory–2 (BFI-2-XS; English version: Soto & John, 2017; German version: Rammstedt et al., 2020)

(b)    the Impulsive Behavior Short Scale–8  (I-8; Groskurth et al., 2022; German version: Skala Impulsives Verhalten-8; Kovaleva et al., 2014a)

(c)     the Optimism–Pessimism Short Scale–2 (SOP2; Nießen, Groskurth, et al., 2022  German version: Optimismus-Pessimismus-2; Kemper, Beierlein, Kovaleva, et al., 2014)

(d)     the General Self-Efficacy Short Scale–3 (GSE-3; Doll et al., 2021; German version: Allgemeine Selbstwirksamkeit Kurzskala; ASKU; Beierlein et al., 2014)

(e)     the Rosenberg Self-Esteem Scale (RSES; English version: Rosenberg, 2014; German version: von Collani & Herzberg, 2003)

(f)     the Internal–External Locus of Control Short Scale–4 (IE-4; Nießen, Schmidt et al., 2022 ; German version: Internale-Externale-Kontrollüberzeugung–4; Kovaleva et al., 2014b)

(g)     the Social Desirability–Gamma Short Scale (KSE-G; Nießen et al., 2019; German version: Soziale Erwünschtheit–Gamma; Kemper, Beierlein, Bensch, et al., 2014)

(h)     the General Life Satisfaction Short Scale (L-1; Nießen et al., 2020; German version: Kurzskala zur Erfassung der Allgemeinen Lebenszufriedenheit; Beierlein, Kovaleva, László, et al., 2015)

(i)     the single-item question used in the European Social Survey (ESS, 2016) to measure self-reported general health

In addition, a set of sociodemographic variables (gender, age, highest level of education, income, and employment status) was measured.

Item analyses

Because R-1 is a single-item measure, item analyses are not applicable. The analysis code for all the other analyses was run with R and can be found in the Appendix.

Item parameters

Table 3 shows the mean, standard deviation, skewness, and kurtosis for the one R-1 item, separately for the English and German sample. The mean showed to be slightly different in both nations and the distribution in the UK showed to be slightly more left-skewed.

Table 3

Descriptive Statistics for the R-1 Item

Note. The rating scale ranged from 1 to 7. UK = United Kingdom (N = 468); DE = Germany (N = 474).

To validate the English-language adaptation of R-1 and to investigate its comparability with the German-language source version, we analyzed psychometric criteria—objectivity, reliability, and validity—in both language versions.

Objectivity

A scale can be regarded as objective when it works (a) independently of the administrator (objectivity of application); (b) independently of the evaluator of the test (objectivity of evaluation); and (c) when unambiguous and user-independent rules are provided (objectivity of interpretation). The standardized questionnaire format and written instructions, the fixed scoring rules and labeled categories, and the reference ranges ensured the objectivity of the application, evaluation, and interpretation of R-1.

Reliability

As estimate for the reliability of R-1, we computed the test–retest stability over a period of 15 and 31 days, with a median of 28 days in the UK and of 20 days in Germany. Our reasoning was that this time span of 2 to 4 weeks is long enough to allow for meaningful test–retest stability estimates while being short enough to preclude the occurrence of pronounced and systematic change in the true scores of risk proneness. The resulting reliability is best understood as lower-bound estimate, as the test–retest stability is sensitive not only to measurement error but also to state fluctuations in risk proneness.

As Table 4 shows, the test–retest reliability estimate for R-1 was .76 (UK) and .83 (DE), which can be deemed sufficient for most research purposes (Aiken & Groth-Marnat, 2006; Kemper et al., 2019). In detail, R-1 proved to be slightly more reliable in Germany than in the UK.

Table 4

Reliability Estimates for R-1

Note. UK = United Kingdom (N = 468; retest: N = 111); DE = Germany (N = 474; retest: N = 117); CI = confidence interval. The time interval between test and retest ranged between 15 and 31 days (Mdn_UK = 28 days; Mdn_DE = 20 days).

Validity

Besides content-related validity, which was shown by Beierlein, Kovaleva, Kemper, et al. (2015) during the original scale development process, we investigated validation evidence based on the relationship between scores on the scale and on other variables.

Evidence based on the relationship between scores on R-1 and on other variables was computed based on fallible manifest indicators. Therefore, the reported correlations are subject to attenuation and represent the lower bound of the true associations. The correlation coefficients are depicted in Table 5; their interpretation is based on Cohen (1992): small effect (r ≥ .10), medium effect (r ≥ .30), and strong effect (r ≥ .50). Due to alpha accumulation through multiple testing, only coefficients with a significance level above p < .001 are interpreted (Table 5 displays unadjusted p values). Before computing the correlations, we recoded the health variable (for both language versions), the “minimizing negative qualities” subdimension of socially desirable responding (for both language versions), and the self-esteem scale (UK only) so that high values represented high self-esteem, high socially desirable responding, and high health values, respectively. In addition, we recoded the employment status variable and tested four different categories against the reference category employed: (a) self-employed vs. employed; (b) unemployed (out of work and looking for work/out of work but not currently looking for work) vs. employed/self-employed; (c) retired/doing housework vs. employed/self-employed; and (d) pupil/student/apprentice/internship vs. employed/self-employed.

In order to investigate this type of evidence, we correlated R-1 with the following constructs: (a) the Big Five dimensions Extraversion, Agreeableness, Conscientiousness, Emotional Stability, and Openness assessed with the BFI-2-XS (Rammstedt et al., 2020; Soto & John, 2017); (b) four aspects of impulsive behavior assessed with I-8 (Kovaleva et al., 2014a; Groskurth et al., 2022); (c) optimism and pessimism assessed with SOP2 (Kemper, Beierlein, Kovaleva, et al., 2014; Nießen, Groskurth, et al., 2022); (d) general self-efficacy assessed with GSE-3 (Doll et al., 2021)/ASKU (Beierlein et al., 2014); (e) self-esteem assessed with the RSES (Rosenberg, 2014; von Collani & Herzberg, 2003); (f) internal locus of control (an individual’s belief that an event is dependent on his own behavior or stable personality characteristics; Rotter, 1966) and external locus of control (an individual’s belief that an event is the result of luck, chance, fate or under the control of powerful others; Rotter, 1966) assessed with IE-4 (Kovaleva et al., 2014b; Nießen, Schmidt, et al., 2022); (g) two aspects of socially desirable responding (exaggerating positive qualities and minimizing negative qualities) assessed with the KSE-G (Kemper, Beierlein, Bensch, et al., 2014; Nießen et al., 2019); (h) life satisfaction assessed with L-1 (Beierlein, Kovaleva, László, et al., 2015; Nießen et al., 2020); and (i) self-reported general health assessed with the single-item question used in the ESS (2016).

Correlations with convergent and discriminant constructs

In both nations, we found with .73 the strongest positive correlation with the subdimension “sensation seeking” of impulsive behavior (see also e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Breivik et al., 2017; Kovaleva et al., 2014a). The second-highest correlation, which was again the same across both nations, was a medium-sized positive association between risk proneness and Extraversion (see also e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Kovaleva et al., 2014a; Tok, 2011). These finding are not surprising because risk proneness is also part of the personality characteristic sensation seeking (e.g., Zuckerman, 2007) and can be considered as part of the facet “excitement seeking” of the Big Five dimension Extraversion (e.g., Whiteside & Lynam, 2001).

Further small-to-medium-sized positive relations we found were with the subdimension “urgency” of impulsive behavior (see also e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Kovaleva et al., 2014a). Risk proneness could be distinguished from the other two subdimensions of impulsive behavior (“premeditation” and “perseverance”) because there were only zero correlations (see also e.g., Beierlein, Kovaleva, Kemper, et al., 2015). We also found a negative medium association with the subdimension “minimizing negative qualities” of socially desirable responding (UK only, with a tendency to the same effect for the German sample). Minimizing negative qualities depicts the impression management component of communion-induced socially desirable responding (Nießen et al., 2019).

With respect to the other constructs, we also found substantial small positive associations between risk proneness and general self-efficacy, Openness to Experience, Emotional Stability, optimism, life satisfaction (UK only, with a tendency to the same direction for the German sample), and health (UK only, with a tendency to the same direction for the German sample). This is in line with previous findings that individuals high in general self-efficacy (e.g., Beierlein et al., 2014; Beierlein, Kovaleva, Kemper, et al., 2015; Olivari et al., 2018), Big Five Openness and Emotional Stability (e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Kovaleva et al., 2014a; Tok, 2011), optimism (e.g., Barel, 2019; Beierlein, Kovaleva, Kemper, et al., 2015), and who report a high life satisfaction (e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Dohmen et al., 2011; Muffels & Headey, 2013) and general health status (e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Dohmen et al., 2011) have a high propensity for also reporting high risk proneness.

Regarding locus of control, previous research suggested correlations between risk proneness and internal locus of control but zero correlations between risk proneness and external locus of control (e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Kovaleva et al., 2014a; Kovaleva et al., 2014b; Lemarié et al., 2019). We found the same pattern for the German sample but in the UK, there were small positive associations with both internal and external locus of control.

Concerning self-esteem, we could not support the findings from Beierlein, Kovaleva, Kemper, et al. (2015), who reported a small positive relation to risk proneness. We found zero correlations instead.

Overall, the pattern of correlations suggests that the evidence based on scale–construct relationships of R-1 was similar to that of its German source version.

Correlations with sociodemographic characteristics

We calculated correlations between R-1 and relevant sociodemographic characteristics, namely employment status, income, educational level, age, and gender. In the present analyses, consistent across the two nations, we found medium negative correlations with age. This finding is in line with evidence from Beierlein, Kovaleva, Kemper, et al. (2015) and Dohmen et al. (2011) that the tendency to higher self-reported risk proneness decreases with age. For the UK, there was a small negative association with gender and retirement/doing housework with a tendency towards the same directions in Germany. Males (see also e.g., Beierlein, Kovaleva, Kemper, et al., 2015; Byrnes et al., 1999; Harris et al., 2006; Dohmen et al., 2011; Weber et al., 2002) and retired people or those who were doing homework had a higher propensity for high risk proneness than females or employed individuals. Moreover, we found a small positive relation between risk proneness and income in the UK indicating that individuals with higher income also reported higher risk proneness. This is consistent with findings from Beierlein, Kovaleva, Kemper, et al., 2015.

Table 5

Correlations of R-1 with Relevant Variables

Note. UK = United Kingdom (N = 468; N_{Employment status} = 450; N_Income = 431); DE = Germany (N = 474; N_Self-esteem = 473; N_{Employment status} = 462; N_Income = 449); CI = confidence interval. Optimism–pessimism: very pessimistic (1) – very optimistic (7). Health: very bad (1) – very good (5). Gender: 1 = male, 2 = female.
^*p < .05, ^**p < .01, ^***p < .001.

Descriptive statistics

Table 6 provides the reference ranges in terms of means, standard deviations, skewness, and kurtosis of the R-1 scale scores for the total population, as well as separately for gender and age groups in both nations. Standard values are not available.

Table 6

Reference Ranges of the R-1 Scale Scores for the Total Population and Separately for Gender and Age Groups

Note. UK = United Kingdom (N = 468); DE = Germany (N = 474).

Further quality criteria

Owing to its good psychometric properties (reliability and validity) and its very short completion time (< 10 s), the instrument can be seen as highly economical.

Acknowledgement

We would like to thank Melanie Partsch (GESIS – Leibniz Institute for the Social Science) for preparing the data.