Risk Aversion 
*************

**Summary**

The global and dimension-specific measurement of risk aversion consists of seven items and was used 
in the SOEP in 2004 and 2009. An individual item on general risk aversion has been included in the 
SOEP at two-year intervals since 2004 and annually since 2008. 

**Theoretical Background**

Risks and uncertainties play a role in almost all important life decisions. In their pioneering work on 
the subject, Kahneman and Tversky (1979) developed a theoretical foundation for the analysis of risk 
attitudes in the form of “Prospect Theory.” This theory describes the process of decision making in 
situations of uncertainty, and postulates that the consequences of a decision are evaluated relative to a 
reference point as either gains or losses, also considering the probability of their occurrence. Here, 
individual risk attitudes affect not only how individuals evaluate the potential gains and losses but also 
how they evaluate the respective probabilities of the event’s occurrence. In the economic research, 
individual risk attitudes are used to explain decisions about financial investments, labor market 
behavior and success, and health behavior (e.g., Dohmen et al., 2011). More recent work in the field of 
educational sociology has studied the influence of individual risk aversion on educational decisions 
(Hartlaub & Schneider, 2013). 

**Scale Development**

The scale was first piloted in the 2003 SOEP pretest and was then used in the 2004 main SOEP 
survey. Simultaneously to the SOEP pretest, a behaviorally oriented experiment on risk attitudes was 
carried out with respondents (see Dohmen et al., 2011). In this experiment, respondents were shown a 
table consisting of 20 text lines and were given the task of deciding in each line whether they would 
prefer to receive a fixed amount of money or take part in a lottery with a 50% chance of winning 300 
euros. In the case of the “safe” option, the amount paid increased progressively from one line to the 
next (0 euros in the first, 10 euros in the second, and so on up to 190 euros in line 20). Since the 
expected lottery winnings would be 150 euros, more risk-averse respondents should take the safe 
option below this value, while risk-seeking participants can be expected to decide for the lottery even 
with safe options of 160, 170, 180, or 190 euros. The amount paid for the safe option at the time of 
switching to this option was regressed on the value of the questionnaire on risk aversion. Here, the 
behavior in the experiment could be predicted well based on the survey responses (β = .61, p < .001). 

**References**

*Dohmen, T., Falk, A., Huffman, D., Sunde, U., Schupp, J., & Wagner, G. G. (2011). Individual Risk 
Attitudes: Measurement, Determinants and Behavioral Consequences. Journal of the European 
Economic Association, 9, 522-550.*

*Hartlaub, V. & Schneider, T. (2013). Educational choice and risk preferences: How important is 
relative vs. individual risk preference? Manuscript submitted for publication.).*

*Kahneman, D. & Tversky, A. (1979). Prospect Theory: An analysis of decision under risk. 
Econometrica: Journal of the Econometric Society, 47, 263-291.*

**Items**

How do you see yourself (Wie schätzen Sie sich persönlich ein): 

Risk Aversion in General 

	1. Are you generally a person who is fully prepared to take risks or do you try to avoid taking risks? (Sind Sie im Allgemeinen ein risikobereiter Mensch oder versuchen Sie, Risiken zu vermeiden?) 

Scale: 0 (Risk averse / Gar nicht risikobereit) to 10 (Fully prepared to take risks / Sehr risikobereit) 

**Test-Retest Correlations**

Risk aversion in general was included in retests taken by subsamples in 2005, 2006, and 2009 within 
30 to 49 days after the respective initial tests. Test-retest correlation pooled across all three waves 
(N = 607) was .60. 

**Items and Scale Statistics**

.. csv-table::
   :file: csv/22a_riskaversion.csv
   :header-rows: 1
   :class: longtable
   :widths: 2 3 2 2 2
   
.. csv-table::
   :file: csv/22b_riskaversion_jugend.csv
   :header-rows: 1
   :class: longtable
   :widths: 2 3 2 2 2
   
.. csv-table::
   :file: csv/22c_riskaversion_bioage14.csv
   :header-rows: 1
   :class: longtable
   :widths: 2 3 2 2 2

.. csv-table::
   :file: csv/22d_riskaversion_bioage12.csv
   :header-rows: 1
   :class: longtable
   :widths: 2 3 2 2 2

**Items**

How would you rate your willingness to take risks in the following areas? How is it ... (Wie würden 
Sie Ihre Risikobereitschaft in Bezug auf die folgenden Bereiche einschätzen? Wie ist das ...): 

Risk Aversion in different Domains 

	1. While driving (beim Autofahren)? 
	2. In financial matters (bei Geldanlagen)? 
	3. During leisure and sport (bei Freizeit und Sport)? 
	4. In your occupation (bei Ihrer beruflichen Karriere)? 
	5. With your health (bei Ihrer Gesundheit)? 
	6. Your faith in other people (bei Vertrauen in fremde Menschen)? 

Scale: 0 (Risk averse / Gar nicht risikobereit) to 10 (Fully prepared to take risks / Sehr risikobereit) 

**Test-Retest Correlations**

Risk aversion in different domains was included in a retest of a subsample (N = 120) in 2009 within 30 
to 49 days after the respective initial tests. Test-retest correlation of the items were (in scale order) .67, 
.48, .58, .69, .51, and .37; scale scores had a test-retest correlation of .69. 

**Items and Scale Statistics**

.. csv-table::
   :file: csv/22e_riskaversionindiffdomains.csv
   :header-rows: 1
   :class: longtable
   :widths: 2 2 2 1 1 2 1