6 “Unobtrusive Methods” (Symbaluk & Hall, 2024)

A major confounding factor in social research is eliminated in nonreaction studies because the observed person or group is not able to react to the measurement process— that is, is not able to manipulate presentation-of-self for the researcher.

— Winston Jackson & Norine Verberg, 2007, p. 143

INTRODUCTION

Most methods rely upon participants who provide the data of interest to researchers through their actions or words, such as experiments, surveys, ethnographic studies, or qualitative interviews. These methods are considered obtrusive because they necessitate an intrusion into the lives of participants to obtain the data. They are also considered to be reactive because participants are responding or reacting to the research instrument (e.g., the questions posed on a questionnaire or an experimental manipulation), the researcher (e.g., a qualitative interviewer), and/or to the study itself (e.g., demand characteristics). In contrast, non-reactive methods, called unobtrusive research methods, are ones in which “the researcher examines evidence of people’s behaviour or attitudes, rather than interacting directly with those being studied” (Strand & Weiss, 2005, p. 161). This evidence takes the form of a variety of sources of information created by and for people (e.g., government statistics, official documents, newspaper stories, personal diaries, song lyrics, and television scripts), and it can also include things left behind by people (e.g., litter, belongings, patterns of erosion). Notably, the information is originally produced for a particular purpose (e.g., song lyrics might be designed in order to express an artist’s creativity and to make money for producers), but it is later examined by social researchers for a different purpose (e.g., to determine the prevalence of gender stereotypes in a music genre), thereby circumventing issues of reactivity since the song is already written and produced.

As early as 1966, Gene Webb and Don Campbell (joined by Richard Schwartz and Lee Sechrest in 1981) wrote about the need for innovative, non-conventional means for finding things out, which they originally dubbed “oddball research” and “oddball measures.” For example, they noted how the careful examination of children’s drawings, library withdrawals, and worn floor tiles could be used to inform researchers about children’s interests, the potential effects of new forms of media, and the popularity of museum exhibits (Webb et al., 2000). Unobtrusive methods include any number of non-reactive measures or techniques. This chapter examines four types of unobtrusive measures that are classified based on characteristics of the data studied: (1) forms of physical evidence examined using physical trace analysis; (2) public and private records that are observed using archival analysis; (3) written, spoken, or visual messages examined using a technique called content analysis; and (4) official statistics and other forms of existing data studied through secondary analysis. This chapter explains how physical trace analysis, archival analysis, content analysis, and secondary analysis can be used to examine data unobtrusively; that is, in a manner that is independent from the original processes that produced it (Lune & Berg, 2021).

PHYSICAL TRACE ANALYSIS

Physical traces are the “remnants, fragments, and products of past behaviour” (Shaughnessy et al., 2015, p. 107). Just as a criminal investigator might examine a crime scene for remaining clues such as finger prints, blood stains, or clothing fibres, researchers can examine locations for physical evidence that helps them to better understand humans based on their past behaviour. This is not unlike what archeologists and anthropologists have been doing for centuries as they study bones, artifacts, and other aspects of material culture left behind by humans to learn more about earlier civilizations. […]

Garbology

The term garbology is frequently used today to refer to the study of behaviour based on the analysis of waste. Garbology as an academic pursuit can be traced to William Rathje (1945–2012), an archeologist at the University of Arizona who founded the Garbage Project, which lasted from 1973 to 2005. In an earlier independent studies project, two of Rathje’s students examined the correspondence between stereotypes and “physical realities” through a comparison of the trash collected from two affluent and two poor families (Humes, 2012). Results showed that all four homes consumed similarly, with the same amounts and types of foods and drinks. However, the poorer households spent more on educational items for their children and they bought more household cleaners than the affluent ones (Rathje & Murphy, 2001). Although questionable due to the small sample size and limited time frame for analysis, the findings nevertheless sparked an interest in Rathje that resulted in a career-long exploration into the assumptions and potential misconceptions related more generally to consumption and waste disposal.

Instead of going out on archeological “digs,” Rathje and his students excavated carefully sampled garbage that was re-routed by the sanitation department and dumped at the University of Arizona (Humes, 2012). For more than three decades, Rathje and his students surveyed, itemized, counted, and weighed garbage, dispelling popular myths about what is most commonly thrown away and what happens to trash after it gets thrown away. For example, while the American public believed that the most common forms of solid waste were diapers and take-out food containers, more prevalent forms of waste turned out to be newspaper and building materials. Another interesting finding was that consumption increased, as opposed to decreased, during food shortages, especially when red meat was scarce. The Garbage Project also showed how many forms of organic waste were unexpectedly preserved by plastics rather than broken down over time in landfills. Finally, how people behave is very different from how they claim to behave, as evident in the under-reporting of unhealthy consumption (e.g., chips, bacon, and alcoholic beverages) and the over-reporting of healthy intake (e.g., cottage cheese, high-fibre cereals, and skim milk) (Rathje & Murphy, 2001).

Graffiti

Graffiti is a special kind of accretion measure that takes various forms and is interpreted in just as many ways. For example, Decker and Curry (2024) define graffiti as:

[A] form of visual communication, usually illegal, involving the unauthorized marking of public space by an individual or group. Although the common image of graffiti is a stylistic symbol or phrase spray-painted on a wall by a member of a street gang, some graffiti is not gang-related. Graffiti can be understood as antisocial behaviour performed in order to gain attention or as a form of thrill seeking, but it can also be understood as an expressive art form. (para. 1)

As a means of self-expression, graffiti commonly takes the form of “tagging,” which involves personalized signatures, or “piecing” and “bombing,” which includes much more highly specialized forms of artwork created with spray cans to embody names and symbolize subcultures (Alonso, 1998). Graffiti also consists of written phrases and statements that express viewpoints and positions (Abel & Buckley, 1977). For example, predominant themes include gang graffiti denoting territories, political graffiti, or derogatory forms of graffiti directed at groups (Alonso, 1998). Graffiti can also take the form of ongoing dialogue as one graffiti writer responds to what another has written, as in the case of bathroom graffiti. The graffiti that appears on the stalls inside bathrooms is usually referred to as “latrinalia” (Abel & Buckley, 1997; Melhorn & Romig, 1985). Smith and Starcke (2023) examined longitudinal patterns of graffiti on campus at a large university in the United States and found a wide variety of forms including themes of awareness, graphic art, school spirit, current events, religion, the environment and sexual orientation. Some forms of graffiti were more likely to be visible untouched for a greater duration than others. For example, religious artifacts tended to last longer than ones related to sexual orientation or school spirit (Smith & Starcke, 2023).

Graffiti also occurs within subcultures and in response to historical events. Rawlinson and Farrell (2010), for example, examined construction site graffiti at 10 large-scale construction projects and identified various common subcultures of graffiti producers (e.g., subcontractors, trades, and football fans), as well as various predominant themes within the graffiti itself, such as celebrations of ownership and personal immortalization. Hagan et al. (2005) studied disaster graffiti following a major flood, which they coined “catastroffiti.” The Red River Valley flood of 1997 took place along the Red River in North Dakota, Minnesota, and southern Manitoba. The most dominant theme in the graffiti following the flood was humour as a coping mechanism. Even within the humorous graffiti, several themes emerged, including jocular humour (e.g., “1-800-big-mess”), sarcasm (e.g., “Are we having fun yet? Keep it up!”), and satirical humour (e.g., “49 FEET MY ASS”) in response to a nearby dike that was supposed to be fortified to 52 feet (Hagan et al., 2005).

[…]

ARCHIVAL ANALYSIS

Archives are historical documents, records, or collections of information that detail the activities of businesses, agencies, institutions, governments, groups, or individuals. To help you understand what archives consist of, consider the various organizations or institutions that can be considered archival sources for information pertaining to you, such as the current school you attend as well as any previous school you attended, employers, creditors, doctors’ offices, and perhaps even a hospital or police agency. Now think of the different ways that information about you and others can be collected and stored, such as in files, as journal entries, and in electronic format. Finally, consider that archival data can be public or private in nature.

Public and Private Archives

Public archives are public records that are prepared specifically to be examined by others. Public archives tend to be continuous and ongoing, as in the case of running records, such as monthly budgets or annual reports; official documents, such as school attendance records or hospital intake records; and library collections, such as periodicals and books. Public records are also prepared, collected, or organized in standard ways, such as arranged alphabetically, organized by date, indexed, and so on.

Researchers sometimes further categorize public archival data based on characteristics of the data itself. For example, Lune and Berg (2021) modernized the original archival categories first described by Webb and his colleagues into three main groups: commercial media accounts, actuarial records, and official documentary records. Commercial media accounts are artifacts created for mass consumption (Lune & Berg, 2021). These public archives can include books, newspapers, periodicals or magazines, television program transcripts, film scripts, photographs, music lyrics, radio broadcasts, X tweets, YouTube videos, website information, and so on. Actuarial records are created for access by a more limited audience but they are technically available to the public under special circumstances (Lune & Berg, 2021). Examples of actuarial records include birth, death, marriage, and divorce records, as well as things like land titles and credit histories. Official documentary records are those produced for a very limited audience although they may eventually end up in the public domain (Lune & Berg, 2021). Official documentary records can include school, hospital, dentist, or medical records, as well as court documents, police records, minutes from meetings, annual reports, sales records, and so on.

Researchers may later examine public archives, such as official records collected for one purpose, for a completely different purpose. For example, Moulden et al. (2010), from the University of Ottawa in Ontario, examined official documentary records collected on convicted sexual offenders to learn more about patterns of sexual offending involving teachers. Specifically, the researchers analyzed archival Violent Crime Linkage Analysis System reports obtained from the Royal Canadian Mounted Police to describe the offending of 113 Canadian teachers who committed a sexual offence between 1995 and 2002. Archival analysis indicated that most offenders abused a position of trust to initiate contact with the victim (84 percent). In addition, many of the offenders befriended the victim (40 percent), offered a form of assistance to the victim (16 percent), or offered the victim money, treats, toys, or work (14 percent). In addition, findings showed that many of the offences occurred at school (44 percent), the offender’s residence (41 percent), the victim’s residence (19 percent), in a religious facility (14 percent), or in another residence (12 percent).

Private archives, in contrast, are personal records created for use mainly by the originating author. These are sometimes directed at a narrow known target, such a friend or loved one. Examples of private archives include personal items such as diaries, letters, or journal entries. The internet is now changing what would traditionally be considered a private archive into what is probably better considered a social domain personal archive. For example, some people post videos depicting their (otherwise) private, personal events on YouTube, while others post highly personal information on social media platforms such as Facebook or Instagram.

[…]

SECONDARY ANALYSIS

Secondary analysis involves the examination of data originally collected by someone other than the researcher for a different purpose. Secondary data can be quantitative or qualitative in nature, as in the case of archived survey data or interview transcripts.

Locating Secondary Sources

Examples of existing Canadian data include official census data and data obtained through national surveys, such as the General Social Survey and the National Population Health Survey. Various government departments and agencies (e.g., Environment Canada, the Correctional Service of Canada), and institutes (e.g., the Canadian Statistical Sciences Institute and the Institute for Social Research at York University) also collect a wealth of information.

Statistics Canada as a Leading Data Source

Statistics Canada regularly collects data on virtually all aspects of Canadian life of interest to federal government agencies. For example, pertaining to criminal justice, Statistics Canada conducts various annual surveys such as the Adult Correctional Services Survey, the Integrated Criminal Court Survey, and the Uniform Crime Reporting Survey (Statistics Canada, 2024d). The Adult Correctional Services Survey collects data from all institutions in the provincial and federal correctional systems on new admissions, conditional releases, and financial/human resources. The Integrated Criminal Court Survey collects information on criminal court appearances and charges. The Uniform Crime Reporting Survey is managed by the Canadian Centre for Justice Statistics, a division of Statistics Canada that is responsible for collecting national data on crime (Statistics Canada, 2023c).

Police agencies throughout the country use the Uniform Crime Reporting Survey (UCR) to report crime statistics in a standard way for criminal offences. Since 1962, the UCR has collected information on the number of criminal incidents, the clearance status of those incidents, and the persons charged. In 1988, a second survey was created (called the UCR2) to obtain additional details on the characteristics of incidents, victims, and accused (Statistics Canada, 2023c). The UCR and UCR2 provide available data on close to 100 different offences. See table 8.2 for summary information on the number and type of homicides committed in Canada between 2018 and 2022. Crime data can be re-examined to test theories, look for patterns and trends, and explore relationships among variables.

It is important to note that Canadian crime data is the end result of crimes that were recorded as such by the police, as opposed to a “true” measure of the amount of crime committed, since many crimes go undetected, many go unreported, and many are not classified as crimes after being reported. Furthermore, there are concerns about how particular crimes are counted. For example, in the case of multiple offences committed at the same time, only the most serious offence is recorded. In addition, the data obtained by police may be shaped by other factors, such as policy decisions, policing practices, and public concerns (Evans, 2020). To try to gain a more accurate estimate of the true amount of crime, criminologists and other interested parties typically supplement official crime statistics with other measures, such as information obtained on surveys completed by offenders and/or victims of crime.

[…]

Activity: Secondary Analysis

VIRTUAL RESEARCH AND DIGITAL RESEARCH

Given the limitlessness of the internet today, this chapter concludes with a brief discussion of internet-based research methods and data that may fall within an unobtrusive or reactive realm, that may include public or private sources of information, and that may or may not require prior ethical approval. The internet has dramatically changed the way researchers conceptualize the social world, and it continues to shape the way social researchers conduct research. First, the internet is a “social space” that offers an endless array of social opportunities and interactions that have meaning for individuals who create virtual selves, participate in online communities, and engage with social media platforms (Hine, 2013). Historically, social researchers have employed traditional methods such as ethnographic fieldwork in online environments (called virtual ethnography) to study individuals or virtual selves created by individuals (Hine, 2000; Pink, 2001). The internet also includes digital technologies that expand the reach of traditional methods into virtual domains such as virtual experiments, online focus groups, or interviews conducted using email. Moreover, the internet contains a wealth of primary and secondary visual data sources as objects of study, in the form of pictures, videos, and graphics that enable researchers across disciplines to examine cultural artifacts, social issues, interactions, and consumption patterns as they exist online (Ardévol, 2012). Networking services such as X or Facebook have amassed such popularity that their “big data” sources have been mined for myriad purposes and have grown beyond what can be adequately examined with more traditional research methods. Consequently, the internet has responded by making available digital tools for examining social issues as digital data, using even more specialized technology. For example, software such as DiscoverText can be used to help sort and analyze textual data in the form of public comments and statements made in blogs and posted to sites such as X (Discovertext, 2024).

Appreciating that the internet can be used in various ways as a source of data or as a facilitator for carrying out research, it should be becoming apparent that previous distinctions made between reactive and non-reactive forms of research, and between private versus public forms of information, begin to blur. Are an individual LinkedIn user’s posts considered private information, shared information among a social network, or a form of primary data that exists in the public domain? Given one’s ability to alter a virtual self, is an online representation best considered reactive or non-reactive? Even the distinction between digital forms of information as data sources versus data gathering methods is difficult to maintain if we consider digital media as the framework for both, as captured by Sarah Quinton and Nina Reynold’s (2018) notion of digital media, which “encompasses all computer-mediated internet and digitally enabled media through which data may be collected, shared and analyzed, including, for example, blogs, online forums, QR codes, online questionnaires, emails, Skype interviewing, YouTube material, Instagram messages, [X] content, geo-location and internet navigation” (p. 10). Alongside the internet’s potential for ever-increasing research opportunities, the ethical challenges stemming from the use of digital media are similarly expansive, particularly in terms of potential risks to individuals and privacy considerations, and ethical issues arise at all stages of the research process, from the nature of the research and its aims, to how the data is conceptualized, to whether and how participants are involved in the research, to how the data is collected and analyzed, and to what is done with the data and findings that result from the study (e.g., see Quinton & Reynolds, 2018).

CHAPTER SUMMARY

1. Define unobtrusive method and explain what is meant by reactive and non-reactive research methods.
   Unobtrusive methods are strategies in which the researcher examines evidence of people’s behaviour or attitudes rather than interacting directly with those being studied. A reactive method is one that directly involves a research participant, who may react to being in a study based on how questions are asked (e.g., in a survey) or due to the presence of a researcher (e.g., in an experiment). Non-reactive methods are unobtrusive because the data is obtained without participant involvement.
2. Explain what physical trace analysis is used for and differentiate between erosion and accretion measures.
   Physical trace analysis is the study of products of past behaviour that provide insight into patterns and trends. Erosion measures are signs of selective wear and tear that tell us about human use and non-use (e.g., worn paths that denote desired walkways). Accretion measures are product deposits left behind based on prior activity, such as garbage or graffiti.
3. Differentiate between public and private archives and note a main advantage and disadvantage of archival analysis.
   Public archives are public records prepared specifically to be examined by others, such as annual reports or attendance records. Private archives are created for personal use and include letters and diaries. As with all unobtrusive methods, archival measures are beneficial because they are non-reactive. In addition, archival records are economical and lend themselves well to longitudinal analysis. Drawbacks to this method include selective deposit and selective survival biases, as well as a potential lack of informed consent.
4. Define content analysis and outline the steps for conducting a content analysis.
   Content analysis is a repertoire of methods that allow researchers to make inferences from all kinds of verbal, pictorial, symbolic, and communication data. Steps for carrying out a content analysis include clarifying research objectives, identifying relevant archival sources, employing sampling procedures, coding data in accord with conceptual frameworks, and summarizing and reporting on the findings.
5. Explain what secondary analysis of existing data entails, identify key sources for secondary data analysis, and note a main advantage and disadvantage of secondary data analysis.
   Secondary analysis of existing data involves an examination of data originally collected by someone other than the researcher for a different purpose. Key sources for secondary data analysis include official census data and data obtained through national surveys, as well as data collected in research centres and institutes. Statistics Canada is the main federal government source for secondary data on most topics involving Canadians. One advantage of using data from Statistics Canada is that it includes built-in privacy and informed consent safeguards from when the data was collected. However, the data was collected for other purposes and may be incomplete for the purposes of the current study.
6. Define digital media and explain how the internet has changed the nature of social research and the methods used to study it.
   Digital media refers to all computer-mediated internet and digitally enabled media through which data can be collected, shared, and analyzed. The internet both provides sources of information for social research and entails means for studying it. For example, online communities are subcultures studied by ethnographers, and virtual ethnography is an approach used to study online cultures. The internet also contains tools, such as software that aids in the collection and analysis of online data.

RESEARCH REFLECTION

1. Suggest a physical trace measure that could be employed as one of the main variables for each of the following research topics:
   • Students’ use of food service providers on campus
   • Students’ typical commute patterns on campus
   • Students’ food consumption on campus
   • Students’ preferred study space on campus
2. Suggest an appropriate archival or secondary data source for each of the following research topics:
   • Popular vacation spots
   • Fashion trends
   • Crime statistics
   • Canadian documentaries
3. Suppose you wanted to conduct a content analysis on the portrayal of gender in popular general-audience-rated movies.
   1. First, search for and list 10 classic movies that are rated G, for a general audience, and are suitable for younger children (e.g., The Lion King, Toy Story, Finding Nemo, and The Little Mermaid).
   2. Now consider who or what is likely to constitute the main character in one of these movies. If you had to categorize the main character as one of only five different “types,” what would the types consist of to ensure that your categories were mutually exclusive and exhaustive? For example, would all animals of any kind, such as mammals, birds, and fish, be classified together?
   3. Come up with one research question on gender you could examine in a content analysis of popular G-rated movies.
   4. Refer to Smith, S. L., Pieper, K. M., Granados, A., & Choueiti, M. (2010). Assessing gender-related portrayals in top-grossing G-rated films. Sex Roles, 62, 774–786. How were the main characters operationalized in this study?

RESEARCH RESOURCES

1. For a comprehensive resource on social media, see Benson, V. (Ed.). (2023). Handbook of social media in education, consumer behavior, and politics. Elsevier.
2. For a comprehensive overview of content analysis, see Faggiano, M. P. (2022). Content analysis in social research: Study context, avenues of research, and data communication strategies. Brill.
3. For a content analysis based on latent messages in song lyrics, check out de Guzman, A. B. and Laguilles-Villafuerte, S. (2021). Understanding getting and growing older from songs of yesteryears: A latent content analysis. Educational Gerontology, 47(7), 312-323.
4. To learn about ethical considerations in secondary analysis, refer to chapter 22 in Gray, D. E. (2022). Doing research in the real world (5th ed.). Sage.

Original Source here: Research Methods: Exploring the Social World in Canadian Context Copyright © 2024 by Diane Symbaluk & Robyn Hall is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted.