Skip to content

Primary data: Additional considerations

Jump to bottom
ubrady edited this page Jul 1, 2026 · 3 revisions

Primary data defined

Primary data is derived from original data sources, i.e., data that is collected directly by a researcher for a specific research purpose or project. Primary data can be collected in a variety of ways. Some of the more common techniques are interviews, ethnographic or field observations, surveys, and experiments. Source: Encyclopedia of Research Design

Advantages of primary data in the IG context

Collecting primary data provides the opportunity to create a research protocol tailored to probe respondents for the specific IG elements needed to answer the research question. For example, if you are interested in determining how stringent the enforcement of U.S. organic farming rules are, you may design an interview protocol that explicitly asks farmers and implementing government officials to identify the Deontic structure of key behaviors and other operationalizations of formal rule constraints, e.g., monitoring mechanisms.

Primary data collection can also aid in:

  • Greater control over the accuracy and relevancy of the collected data
  • Better alignment with research-specific questions/objectives
  • Identification of institutional statements and syntactic elements that are most relevant for understanding rules-in-use in a given context. Interview protocols can be designed to increase the number of institutional statements and grammatical elements derived from transcripts.

Maximizing the number of statements and/or IG components included in the analysis can smooth out data inconsistencies and outliers, establish a more consistent foundation from which to describe the institutional design, and generate explanatory insights that support generalization and scientific discovery. Strategies to increase the number of institutional statements/syntactic elements include conducting longer, more in-depth interviews and/or interviewing more people.

Data saturation

One strategy for determining when sufficient information has been collected is to use data saturation, that is, continuing data collection until additional interviews or collected data no longer generate substantially new institutional statements or syntactic elements. There are ongoing debates about when saturation is achieved across methodologies and disciplines. We recommend consulting the literature to determine the most appropriate methodology for your work. For a general overview/guidelines, we suggest:

The number of institutional statements/syntactic elements can be increased through targeted probing of interviewees for a variety of syntactic elements, different rule types, and/or by interviewing stakeholders holding a diverse range of positions (e.g., government officials, organic and traditional farmers).

Addressing ambiguities and missing data

For rules-in-use, it is crucial to probe for any ambiguous or missing IG syntax elements. For example, if an interviewee refers to actors using the term “we,” the researcher should confirm whether “we” represents all actors in a given situation, or refers to specific actors (e.g., farmers), while also determining any specific characteristics associated with the actor/actor group (e.g., organic farmers). It is also important to clarify the circumstances, conditions, and constraints of each statement (context), the mandatory or non-mandatory nature of the activity (Deontic), and the consequences of noncompliance (Or else).

While the Or else component in formal rules often includes tangible consequences, such as economic fines, rules-in-use may involve less visible consequences, including internal ones, like emotions, social disapproval, reputational costs, spiritual consequences, or negative self-judgment. These consequences may be difficult for interviewees to recognize or express explicitly. One strategy to facilitate complete responses is to incorporate visual aids in the interview guide, particularly for Deontic elements and consequences. For example, researchers can present index cards listing institutional statements without their associated Deontics or Or else and ask interviewees to add the grammatical elements they associate with each institutional statement (Siddiki et al. 2012). However, caution should be exercised to ensure that interviewees are not inadvertently pressured into providing answers when they are uncertain about an institutional statement, or when a specific grammatical element is not actually present.

During interviews, it is also important to clearly distinguish between statements that describe individual and social behaviors. Individual behaviors are driven by internal motivations, such as personal emotions and individual economic incentives.

Social behaviors refer to behaviors shaped by the social context in which actors are embedded. They may be motivated by a mix of formal rules, internal and external motivations, such as social emotions, status, or the desire to fit in a community. This distinction is important for rules-in-use because researchers may need to determine whether a reported behavior reflects an individual preference or a broader socially shared practice. Asking interviewees about the reason behind a described behavior can help distinguish individual and social behaviors, and tracking answers to the same question across multiple interviews helps confirm whether the behavior reported by individuals reflects a broader pattern of shared understanding within a study group. For example, consider a hypothetical example of a person, who reports leaving money after picking tulips in a field because they feel ethically obligated to do so and because they observed others doing the same. Such a statement might be an indication that leaving money after picking tulips is both an individual and social behavior. However, to confirm that it is a social behavior, it is important to assess whether other tulip pickers report similar reasons for leaving money. See Table 3 for additional example questions.

Distinguishing between regulative and constitutive statements

Finally, most primary data will describe behavior and actions and, as a result, most statements will be regulative in form. However, researchers should also pay attention to constitutive statements that outline system parameters and their corresponding Modal elements, as their violation may threaten the existence of the governance system itself (see Table 3 for examples of questions and Table 1 for Modal element). For example, interviewees involved in species’ Appendix listing debates within the Convention on Migratory Species meeting expressed divergent definitions of migratory species. Since the definition of migratory species represents a core concept in the governance forum, it is a constitutive statement. If CMS decision-makers operate under different migratory species definitions, the potential for conflict and a concomitant unraveling of the governance system is high. Asking interviewees about constitutive statements, like definitions, which positions exist, how these positions are established, the respective authority imbued in select positions, and any other governance system constraints will thus provide more nuanced information on the governance system being analyzed.

Disadvantages using primary data

  • Increased time, effort, and costs associated with data collection
  • Other common pitfalls of human subjects’ research, such as subject and researcher bias, research design flaws, and recruiting challenges

Clone this wiki locally