Technical Background

Technical guide

For users who are interested in the algorithm assumptions and calculations that guide WellWorth, you’ll find  that information here.  

Conversion of wellbeing indicators

Subjective Wellbeing is traditionally measured using a number of indicators and scales, and so do the studies analyzing the link between wellbeing and its co-benefit. It is therefore important to have a mechanism allowing conversion of one indicator into the other, in this way increasing in the potential of our tool. This is obviously the second best solution, as different wellbeing indicators are designed to measure different aspects of wellbeing. Moreover, it has been shown that positive and negative wellbeing indicators are not simply “the opposite ends of a continuum” (Cohen et al. 2003, p. 652) and actually capture different elements of wellbeing. Ideally, we would need the relationship between wellbeing and its co-benefit to be studied using a number of indicators. However, this is seldom the case and most studies focus just on a single or a few wellbeing indicators.

To overcome this problem WellWorth includes a conversion tool, which allows you to translate any wellbeing indicator into life satisfaction points. The conversion tool exploits the results from Mukuria et al. (2015) of the Policy Research Unit in Economic Evaluation of Health and Care Interventions and from the Cross-Cutting Group (2015). These studies take life satisfaction as a reference indicator and, using data drawn from several population surveys in the UK, measure the exchange rate (i.e. correlation) between several wellbeing measures.

Wellbeing indicator STD UNSTD Source
Life satisfaction (0-10) 1 1
Worthwhile (0-10) 0.80 0.80 HIPO
Happiness (0-10) 0.84 0.84 HIPO
Anxiety (0-10) 0.60 0.53 HIPO
ONS-4 total (0-40) 0.91 0.25 HIPO
SWB-VAS (0-10) 0.82 0.49 HIPO
WEMWBS (14-70) 0.68 0.13 SYC65
SWEMWBS (7-35) 0.66 0.26 SYC65
GHQ-12 (0-36) -0.49 -0.21 US
GHQ positive (0-18) -0.40 -0.42 US
GHQ negative (0-18) -0.48 -0.30 US
SF-6D (from SF-12) (0-1) 0.51 9.15 MIC
SF-6D (from SF-36) (0-1) 0.48 9.22 MIC
EQ-5D-5L (-0.6 – 1) 0.63 5.65 HIPO
ICECAP-O (0-1) 0.63 8.77 SYC65
ICECAP-A (0-1) 0.65 9.71 MIC
EQ-VAS (0-10) 0.70 0.80 HIPO

Source: Mukuria et al. (2015) and Cross-Cutting Group (2015)

STD indicates correlation coefficient between Life satisfaction and the other wellbeing measures. UNSTD indicates number of units change in Life satisfaction following a unit increase in the selected wellbeing measure (Cross-Cutting Group elaboration).

Data Sources used by Mukuria et al. (2015): HIPO (Health Improvement and Patient Outcomes); MIP (Multi Instrument Comparison); SYC65 (South Yorkshire Cohort over 65); US (Understanding Society).

Core assumptions behind the tool

Time frames of impact

An important assumption made by WellWorth is that the increase in wellbeing following the wellbeing intervention will persist over the entire time frame of the analysis. This might of course sound unrealistic give that the preferences of individuals are formed in response to their circumstances and are dynamic ie change over time according to the external environment. However, taking into account the dynamic evolution of preferences would involve making more arbitrary assumptions and this would increase WellWorth’s complexity beyond a manageable level.

Default values

In section x, WellWorth provides default values derived from the empirical evidence for the age profile of your chosen population and the employment rates by age group. If you don’t have the relevant information for the population undergoing the wellbeing intervention, you can use the default values instead. .

Discounting

All future monetary values are discounted at the default value of 3.5%, as suggested for periods of 0-30 years by the HM Treasury.

Inflating results

Monetary values derived from empirical studies based on market prices of a certain year, are inflated to 2015 prices using inflation rates based on the Consumer Price Index. Income values are updated to 2015 values according to the increase in wages derived from ONS official statistics based on Annual Survey of Hours and Earnings.

Time frame of analysis in years Years when income becomes affected by LS Discounted present value of age effect on income = 0.05^time*1/(1+r)^time
1 0 1
2 1 1.014492754
3 2 1.029195547
4 3 1.044111425
5 4 1.059243474
6 5 1.074594829
7 6 1.090168667
8 7 1.105968213
9 8 1.121996738
10 9 1.13825756
11 10 1.154754046
12 11 1.171489612
13 12 1.188467723
14 13 1.205691893
15 14 1.223165688
16 15 1.240892727
17 16 1.25887668
18 17 1.277121269
19 18 1.295630273
20 19 1.314407523
21 20 1.333456908
22 21 1.35278237
23 22 1.372387912
24 23 1.392277592
25 24 1.412455528
Cancel