Tasks in this step are likely to fall into two categories:
- Collect data for measurement and valuation
- Analyze and combine measurement and valuation data to obtain valuation results
Collect data for measurement and valuation
Companies may have already identified some of the data sources and the best collection methods in Steps 8 and 9. This information is likely to have contributed to the choice of indicators and valuation techniques; however, it is valuable to review at this stage.
It is equally important to determine when and how data is collected as it is to determine what data is collected. For example, if you are trying to gather information about the impact of an investment in training for farmers it is important to keep in mind the cropping or harvest season. If companies are comparing performance year to year it is important to clarify the baseline, as well as ensure consistency in the timeframe and methods for measurement. This allows for comparability.
It is important to make sure that the right data is going to be collected to support whatever analysis or calculations are required to produce the social capital valuation. If incorrect, or insufficient, data is collected, then time and resources will be wasted by having to fill data gaps or re-run data collection.
The following is an overview of the different data sources that companies might use in the context of the Protocol.
Primary data: Data collected by the company (or externally contracted party) specifically for the assessment. This offers more precise results but can also be more time-consuming and require more specialized skills. Some sources of primary data include:
- Internal business data: this includes data collected by the company, e.g. on hiring and recruitment from human resources, spending with suppliers, or performance of sales teams.
- Surveys: this includes surveys conducted by the company of a specific population such as employees, suppliers, users or customers.
- Interviews or focus groups: this seeks to gather the perspectives of the target population.
Secondary data: Data that was originally collected and published for another purpose or a different assessment. Secondary data sources include:
- Peer reviewed literature: This includes reports or studies prepared by credible research parties.
- Interviews with third-party/proxies: These include interviews with experts such as local NGOs who can provide insight into communities that may be inaccessible to the company.
- External data: This could include government statistics or World Bank/UN databases, such as household budget surveys, demographic health surveys, or other routine data collection databases.
- Existing analyses: This includes data produced from existing calculations or models, which may have been completed internally or from external sources.
Most businesses use a combination of primary and secondary data as this is more practical and can be sufficient to achieve the purpose of the assessment.52
Both primary and secondary data-gathering techniques can be used to collect data beyond a company’s own operations, i.e. upstream or downstream in the value chain. In some cases, the engagement required to collect primary data from upstream suppliers or with downstream customers can be of business benefit in itself due to strengthening of the business relationships. Care should be taken so that data collection requests from suppliers or customers are not overly frequent or onerous. Where possible and in the case of suppliers particularly, feedback should be given to the data provider to help them understand how they compare to their peers, to help them improve performance and to help them see the benefit of their efforts in providing data.
One important factor to be aware of with surveys and focus groups - whether designing primary data collection or when interpreting secondary data - is sampling and sample bias. Instead of collecting information from all individuals or households in the community, companies may select a representative sample, and, based on that sample, produce an estimate of the indicators of interest, which can be generalized to the entire population. Where this is done, care should be taken to ensure the sample shares the same key characteristics as the population is it being used to represent.
Box 5: Ethical considerations in data collection
The Protocol is not intended as a guide to all of the different data-gathering methods. However, there are some important ethical requirements and principles for data collection, especially when engaging directly with communities. This not only ensures respect for the rights of participants but also strengthens the accuracy of the results. Anyone engaged in collecting data from communities should be aware of:
- Informed consent: This is the process of getting approval from participants for the sharing and use of data. It is an ethical requirement for most research. To ensure that consent is informed, consent must be freely given with sufficient information provided on all aspects of participation and data use.
With regards to indigenous communities, companies should abide by specific principles around free prior and informed consent as specified by the UN.
- Cultural norms: Companies should be sensitive, aware, and respectful of cultural norms when determining appropriate data collection techniques. This could include, for example, being conscious of gender dynamics and whether women will speak freely in front of male peers.
- Legal requirements: Companies should review data laws and regulations in the country and locations where they are collecting data to ensure they comply.
Other factors to be aware of include education and literacy levels, privacy and anonymity, as well as safety in some contexts.
Analyze and combine data to produce valuation results
Consideration may already have been given to how the collected data sets are to be combined in order to produce a valuation result during the previous steps. If not, this should be done now. Additional technical support may be required - internally or externally - to complete more complex calculations or modelling.
There are a few considerations companies should keep in mind when conducting valuation:
- Acceptability of valuation approach to stakeholders: As has been discussed in Step 7, it is important to recognize that not all stakeholders will recognize valuation - and in particular monetary valuation - of some impacts or dependencies. For example, whilst it is common in government policy appraisal to value a human life in monetary terms53 some stakeholders may disagree that this is ever appropriate. Whether or not stakeholders are likely to accept the type of approach used to value social capital is a key part of identifying an approach that is fit for purpose (and should be decided upon prior to data collection).
- The ethics of making trade-offs: Just because it is possible to value an impact does not, by itself, justify trading one impact off against another that may have been valued more highly. Similarly, the net value of the impacts from an activity may be positive but there may be negative impacts as well. For example, there may be situations where employment and wage payments create value for workers but working conditions are unfavorable. It is important to look both at the total value and the individual parts to ensure that the company isn’t overlooking any key risks.
- Double Counting: Companies should aim to avoid double counting when conducting valuation (and measurement). This can occur, for example, when intermediate costs and/or benefits, rather than only final costs and/or benefits, are assessed. For example, the value of wheels is included in the price of a car sold. So recording both the price of wheels and the price of cars themselves on a balance sheet is a classic example of double counting.
- Assumptions: Companies should identify and communicate all assumptions used. This can include for example the use of average or minimum wages to capture changes to employee incomes and any adjustments made to include other benefits paid by the company.
It is important to note that the practice of valuing social impacts and dependencies by companies is in its infancy, and its application is still a challenge. Particularly challenging is that different social impacts require tailored approaches to apply values to them and there may be a number of alternatives to choose from for each. Differences between these alternatives may include their level of precision, their granularity and the completeness of the value that they represent. As discussed in Step 7, care should be taken to make sure that any values are consistent with one another - especially if they are to be directly compared or aggregated - and that they are fit for purpose. It is also important to conduct sensitivity analysis, and to be transparent in your level of confidence in the results.
52 Adapted from the Natural Capital Protocol (p.62)
53 See, for example the OECD, Mortality Risk Valuation in Environment, Health and Transport Policies