The Resource Generator (Snijders, 1999; Van der Gaag & Snijders, 2005) is a survey instrument for the measurement of individual social capital. Different from other measurement instruments for social capital, it does not include the mapping of an ego-centered network (as with the use of Name Generators), which may save a lot of interview time. Instead, it bears more similarity to the Position Generator (Lin & Dumin, 1986; Lin, Fu, and Hsung, 2001). Different from the Position Generator however, Resource Generator information directly refers to accessed social resources instead of occupational prestige. Whereas the Position Generator can be used to measure access to social resources useful in instrumental actions, the information retrieved by the Resource Generator can more clearly refer to social resources useful in expressive actions.
strong and weak points
Each social capital measurement method has specific disadvantages: Name Generators can lead to lengthy interviews, and Position Generator data requires an extra interpretation step in terms of the instrumentality of accessing social prestige. A specific difficulty with the Resource Generator is composing the instrument, as its questionnaire items are directly based on theoretical ideas about the usefulness of specific social resource collections for the general individual. A concrete answer to the question "What are useful and usable social resources?" may vary considerably, as social resources differ in their availability, visibility and appropriateness for exchange between countries and populations (Van der Gaag & Snijders, 2005). The construction of a Resource Generator should therefore be pursued with substantive theoretical guidance; else studies may end up with incomparable measures based on arbitrarily selected items, similar to much previous social capital research using name generators (Lin, 2001b). Since you probably want your social capital study to be comparable with other research, it is wise to also include Position Generator items besides Resource Generator items. Some empirical relationships between Resource Generator measures and Position Generator measures are discussed in Van der Gaag, Snijders & Flap (2008).
General ideas for the construction of new Resource Generators are presented in Van der Gaag & Snijders (2004, 2005). In addition, your version could be inspired by the studies listed below and the checklist in this document. Instead of developing a new instrument, it may also be possible to re-use existing, name generated data: each name generator item with at least one alter mentioned in response can also be considered a resource generator item (see Van der Gaag, 2005; chapter 8), and the same is true for each name interpretation item provided this contains information about resources. Such data may be converted into responses 'as if' resulting from Resource Generator items.
Especially researchers in health research settings are sometimes explicitly looking for 'validated' versions of measurement instruments, that may also require payment to be included in studies for copyright reasons. The good news is that although I love to get paid for my work, I think scientific knowledge should be free. The instruments listed below have generally been developed in projects paid for by European tax payers. Should you use that knowledge, no money is required, but a notification of your findings and adjustments is appreciated. Regarding 'validation', I do not (yet) think there is a definitive Resource Generator that could qualify as 'the' validated version. There have been some replications for general populations by now you can use as a template (see below). However, you will have to make your own version for the population you study and validate that version, especially when you are looking to unveil dimensions in social capital, or when you study non-western populations or special subpopulations for which such items are not distinctive enough. (I can help, but not always for free.) Finally, 'validated' tends to suggests fully known measurement properties of an instrument and distribution of its scores over populations. The predictive validity of the Resource Generator and its subdimensions have so far made sense. However, diagnostic value on an individual level, or comparison with norm scores are beyond current knowledge.
The Resource Generator has been used so far in quite a few studies for many different populations. An overview up to 2010 can be found behind this link. Contents of additional, as yet unpublished studies and ideas for sets of items are listed below. If you want to use a Resource Generator in your own research, you are welcome to download the available versions, and adjust them to your needs. Of course, if you eventually compose and use a Resource Generator in a data collection I appreciate to hear from you. Please send me a copy of your version, a short description of your study, and references to written material or an internet link to your research, so this website can stay updated, and others may benefit from your findings.
Social Capital for the long term unemployed - the Netherlands
In 2010, Peter Bos and I discussed a list of social resources potentially helpful to individuals to escape a situation of long term employment. This set of 32 items is partly a modification of the SSND (1999 and 2008, see below) and work by Webber (2007).
downloads: Resource Generator for long-term unemployment[English/Dutch]
Social Capital for Single entrepreneurs (ZZPs) - the Netherlands
Here, the research context of social capital is narrowed to the domain of entrepreneurs. In a study investigating the productivity and goal specificity of social capital for single entreprenerus without personnel (ZZP's) a list of 9 resources was used. In addition, 4 items denoting negative roles in the network were added.
downloads: Resource Generator for social capital of single entrepreneurs without personnel[English]
Social Capital on the work floor - the Netherlands
Most Resource Generator item lists are designed for general populations. However, for some research questions the context of social capital is preferrably narrowed, for which cases Resource Generator items need to be much more specifically adjusted. In a study investigating productivity and goal specificity of social capital for individuals working in organizations, a list of 34 resources was draw up deemed representative for most organizations. In addition, 7 items denoting negative influences or 'resources' from the network were added, since so relatively few studies investigate these systematically incorporated into a social capital perspective.
downloads: Resource Generator for social capital on the work floor[English and Dutch]
SSND - the Netherlands
Originally, the Resource Generator was designed for the Social Survey on the Networks of the Dutch (1999-2000), a nationwide study of social network content in general, and its role in neighbourhoods and on the work floor in specific. The study used a representative population sample (N=1,007), and is the basic material for my writings about the measurement of social capital for individuals. The SSND Resource Generator aims to measure general social capital in a population, and includes 33 items to enable dimensional analyses (see Van der Gaag & Snijders, 2003, 2004, 2005). Note that this version is designed for administration during a face to face-interview. In 2007 a second wave of SSND study data was collected, which also included improvements to Resource Generator items suggested by earlier findings.
downloads: SSND Resource Generator wave 1 [Dutch], SSND Resource Generator, Position Generator and Name Generators wave 1 [English].
SAFIRE - United Kingdom
New versions of the Resource Generator will be used in the UK SAFIRE study: "social factors in recovery from depression", conducted at the Institute of Psychiatry, Kings College, London. Measures from the instrument will also be used as an outcome in a study evaluating a project in London encouraging people with severe mental health problems to become volunteers (Capital Volunteering). The 27-item Resource Generator in this study is inspired by the version used in the Dutch SSND, but includes improvements suggested by extensive pre-tests (see Webber, 2003; Webber & Huxley, 2007; further results published in Van der Gaag & Webber, 2008; Dutt & Webber, 2009). It includes item content adjustments based on expert panel evaluation and focus group outcomes. The SAFIRE Resource Generator aims to measure general social capital in a population, and is designed to be self-administered. A copy of the SAFIRE questionnaire can be requested from the author.
CONNECTED LIVES - Toronto, Canada
A short, 7-item Resource Generator will be used in the Connected Lives study - the third study of social networks the Toronto borough of East York (N=900) conducted by the Department of Sociology at the University of Toronto. This version aims to measure general social capital in a population; note that it is designed to be self-administered.
downloads: Connected Lives Resource Generator [English].
In a population study conducted by the Department of Psychology, European Humanities University, Minsk, a 22-item Resource Generator was used which was directly based on the Dutch SSND instrument. The initial distribution of responses to the items, which are clearly different from the Dutch ones, provide an example of how such items may appeal differently between populations.
A special version of the Resource Generator was being developed for the measurement of individual social capital in Sucre, rural Bolivia (Lancee, 2005). More details, among which Spanish versions of the used questionnaires, can be requested from the author.
downloads: Resource Generator for rural Bolivia [English]