Don W Stacks & Marcia Watson DiStaso. 21st Century Communication: A Reference Handbook. Editor: William F Eadie. 2009. Sage Publication.
Public relations research (also known as public relations measurement) is a hot topic. This comes in light of the demand for accountability and transparency in organizations. Campaign decisions are not and should not be made in the dark but instead require research to make informed decisions. Public relations research focuses on the entire public relations process and examines the relationships that exist among and between an organization or person and their target publics (Lindenmann, 2006).
Public relations research is the preparation and gathering of “data” in support of a campaign to inform or change a public’s or audience’s perception or behavior. Public relations research differs slightly from what most would call research in that it has been primarily set in the field in support of business objectives. Academics have argued that its focus needs to shift from descriptive to inferential. That is, public relations research has focused on the here and now, describing what the products of a campaign or program were instead of inferring the impact of those products on the outcome of interest. (A campaign for our purposes is defined as a public relations effort that has specified beginning and ending points. A program, on the other hand, is a public relations effort that once begun continues for an unspecified time span. As such, public relations campaigns are typically associated with agencies, while programs are associated with corporate [internal] public relations efforts.) For instance, a research might describe the number of press releases submitted and printed or video news releases (VNRs) distributed and shown—explaining what was accomplished as a simple description and not providing evidence that the public relations activities actually influenced the final outcome.
Regardless of the argument, research in the practice of public relations differs from academic public relations research in that its focus is often on establishing the impact of public relations on return on investment (ROI). The key to understanding public relations’ impact on ROI is to view public relations as a mediating factor in the introduction, change, or reinforcement of conditions that will affect the outcome of a client’s bottom line. As such, public relations seeks to provide the information, motivation, and behavioral intent necessary to get a specific targeted audience to act (or not act) in a particular way. This has moved public relations from a technical production skill to a strategic managerial position in the corporate environment and a counselor position in the agency environment. Thus, public relations strategy focuses on the effects of actions rather than on the production of products supporting those actions.
The role of public relations research from the early 20th century to date has shifted from the simple act of counting publicity materials to a more sophisticated evaluation of public relations’ effectiveness. This shift, however, still relies primarily on the descriptive nature of the communication as “published” in the mass media, often aimed at establishing a “buzz” for the client or product. This activity has in the late first decade of the 21st century become known as “word-of-mouth” promotion, and its value is seen in the marketing industry’s taking it over as one of its tools.
Today’s public relations research activities, although still descriptively entrenched, have moved toward making inferential predictions of effect. This has resulted in a new breed of practitioner and academic who is well versed in advanced methodology and statistical applications.
History and Theory
Although we can trace the history of public relations to the ancient Egyptians and its more modern practice to the Romans, time has resulted in a chasm between public relations education and public relations practice. In tracing contemporary practice, one goes back to the early-20th-century “pioneers” such as Edward Bernays, Paul Garrett, John W. Hill, Doris Fleishman, Ivy Lee, and Arthur W. Page. Although there were others, Bernays, Fleishman, Hill, and Lee represent one large practice approach, while Paul Garrett (General Motors) and Arthur W. Page (AT&T) represent the other large practice approach. The former were counselors, and from their initial work came the large public relations agencies of today. The latter was the model for what is labeled today as corporate communications. The agency represents a number of clients and products, and agencies range from “full service” to those that are specialized in what they handle (e.g., crisis management, travel tourism, integrated communications), often taking on a marketing orientation. The corporate communication practitioner is usually a corporate vice president or senior vice president whose role is the strategic management of corporate communication to a number of key audiences, often split between stock- and stakeholders, employees, governmental agencies, customers, and financial analysts. It should be noted that most of the early public relations practitioners were trained in the journalist tradition, a training that still has an impact on contemporary public relations research.
Grunig and Hunt (1984) offered a historical approach to 20th-century public relations in the form of public relations functions that can still be seen today. They labeled early-20th-century practice as “press agentry.” That is, public relations was practiced as propaganda and simple promotion. Reaction to propaganda and promotion evolved into a “press information” function, where information was crafted based on more factual and truthful communications aimed at the press. The role of public relations was to get the client’s message to media “gatekeepers” (editors, opinion leaders) through press releases, media alerts, and the like and then track how many times they were picked up in the media.
After World War II, the public relations function became more strategic. Grunig and Hunt (1984) label this function two-way asymmetric; this was where the social scientific method and analysis was first applied to public relations practice. In this model, the effect of messaging was evaluated through survey research, and the campaign and its messages were reevaluated. Today, the model describes public relations as a “two-way symmetrical” function where public relations is practiced with continuous feedback in the communication environment, and the public relations function has moved from technical producers of messages to the strategic decision-making realm of top management. Furthermore, it has been stated that what should be aspired to is excellent public relations that is “both symmetrical and asymmetrical, two-way, ethical, and both mediated and interpersonal” (Grunig, 2001, p. 30).
Although Grunig and Hunt’s model is simplistic, it does reflect the change in public relations from the early 1900s to date. Over the years, there has been one constant—the public. As Grunig and Hunt (1984) said, “If an organization does not need to be responsible to its publics, it also does not need a public relations function” (p. 52).
Another significant change in public relations has come as a response to new technologies. Both the academic study and the practice of public relations, has never been more difficult; this is because the public spaces in which we compete and communicate have never been more complex and unpredictable. In a matter of seconds, organizations have the ability or obligation to communicate directly to countless stakeholder groups about information that directly affects their success. This includes online forums, chat rooms, blogs, social-networking sites such as Second Life, Twitter, and Flickr, along with collaborative Web sites such as Wikipedia, YouTube, MySpace, Flickr, http://Linkedin.com, http://Writely.com, and http://Ma.gnolia.com. These allow for two-way communication in ways that are evolving more and more each day.
Research and the Academy
By definition, one would assume that academic research would be based on theoretical formulations. Public relations academic research, however, has been more informed by the practice than by theory. This is not to say that public relations is atheoretical; indeed, public relations has developed a rich theoretical base, but it is a young academic discipline, one that like other communication areas borrows heavily from other disciplines. The vast majority of public relations research is found in descriptive methodologies and simple counts. Historical case studies, surveys, and qualitative approaches emphasizing in-depth interview and focus group methodology were dominant until the late 1980s, as evidenced by articles in Public Relations Quarterly (PRQ) and Public Relations Journal (PRJ). PRQ and PRJ focused largely on matters of practice. In 1989, the volume that would become the Journal of Public Relations Research, the Public Relations Annual (Vols. 1-3, edited by Larissa and James Grunig) produced the first academically oriented journal in public relations that looked for a theoretical base to research. As will be noted later, some very sophisticated research methodologies have been employed in the academic side of public relations research.
Research and the Practice
When considered in isolation, public relations research can be described as working along a continuum ranging from precampaign activities through the actual campaign implementation to evaluation of campaign effectiveness and finally to how that campaign helped the business (company, client, product) achieve its overall objectives. Research, then, can be assessed by the methods employed in each of the three phases of a campaign and across the three phases and extended to correlate with the business outcomes of interest. We label these phases as developmental, refinement, and evaluation. The correlation to final business objectives can be looked at phase by phase or continually through the campaign.
This correlation between outcomes and phases represents one of four assumptions public relations practitioners make about research. They also assume that decision making is approached the same way in all organizations, that all research is based on measureable objectives from which communication strategy is formed, and that public relations research is behavior driven and knowledge based (that it is formulated in public relations and social science theory). To understand this, you must first understand how business objectives drive research objectives, which, in turn, contribute to ROI.
The goal of public relations research is to meet the larger objectives of a client (person, organization, or brand). As such, public relations research focuses on the perception of the client in three areas: (1) information (awareness, knowledge), (2) motivation (internal reasons for action), and (3) behavior (predictive of actual behaviors). Research objectives must include a clearly stated causal relationship: what is done (“outputs” such as press releases, VNRs, blogs) to influence the messages of opinion leaders (“outtakes” such as editorials, analyst reports, experts) that lead to certain ends (“outcomes” such as purchase behavior, reputation ratings, relationship evaluation, trust). The general notion is that informational objectives must be met prior to motivational objectives, which must be met before behavioral objectives can be met. That is, behavior that is not motivated by key messages is idiosyncratic and random.
To write clear, concise, and measurable objectives, one must be familiar with the differences between outputs, outtakes, and outcomes. According to the Dictionary of Public Relations Measurement and Research (Stacks, 2006, pp. 14-15),
outputs are what is generated as a result of a public relations program or campaign that influences a target audience or public to act or behave in some way—this is deemed important to the researcher (also known as a “judgmental sample”), the final stage of a communication product, production, or process resulting in the production and dissemination of a communication product (brochure, media release, Web site, speech, etc.);
outtakes are the measurement of what audiences have understood and/or heeded and/or responded to in a communication product’s call to seek further information from public relations messages prior to measuring an outcome; audience reaction to the receipt of a communication product, including favorability of the product, recall and retention of the message embedded in the product, and whether the audience heeded or responded to a call for information or action within the message; and
outcomes are quantifiable changes in awareness, knowledge, attitude, opinion, and behavior levels that occur as a result of a public relations program or campaign, an effect, consequence, or impact of a set or program of communication activities or products, and may be either short term (immediate) or long term. (p. 14)
Thus, through the use of properly written objectives, a public relations campaign uses research to establish benchmarks against which to gauge campaign effectiveness, check for progress during the actual campaign, and determine whether or not the campaign contributed to the client’s overall business objectives. Depending on the specific objectives, different types of research methodologies are employed across the campaign.
There are three general types of research methods employed in public relations research. Public relations methods include historical/secondary, qualitative, and quantitative approaches to research. Historical/secondary research is usually found at the developmental stage, where previously published research is reviewed to establish industrylike benchmarks. Sometimes the client’s needs are such that no historical or secondary research is available and the practitioner must conduct primary research of his or her own to establish benchmarks—through in-depth interviews, surveys, focus groups, or even participant observation. Conducting primary research, however, is more expensive than conducting historical or secondary research. Although experimental or simulation methodologies are also available, currently they are rarely conducted in public relations.
During the refinement phase, surveys and polls, in-depth interviews, and focus groups may be employed at regular intervals across the campaign to verify that informational and motivation objectives have been met. If they have not been met, then the outputs associated with campaign strategy are reexamined, and the strategy is altered. Research at the various intervals has targeted goals set against preestablished benchmarks.
Initial research across targeted audiences may first establish that the information being communicated has been seen or heard (information awareness objective) and is understood (information knowledge objective) through a poll or survey or through requests for more information contained within the outputs being used (press release, VNR, paid advertisement, 1-800 call numbers, or mailer cards). While this can ascertain whether large numbers of people are aware of and understand key campaign messaging, more qualitative data may be required through the use of focus groups (moderated group discussions) and possibly in-depth interviews with key influentials. Focus groups and interviews may be used to gather more in-depth information about whether the strategies employed are on target.
If it has been ascertained that the campaign’s informational objectives have been met, then motivational objectives can be tested. Generally, this is done through audience surveys, through either one-shot surveys or longer, more expensive panel surveys where the same participants are questioned over a set length of time. The surveys attempt to gauge attitudes and beliefs toward the campaign’s client and serve as a way of establishing predispositions toward behavior. If attitudes have been modified or maintained (depending on the campaign’s goals), the research then attempts to gauge the behavioral intentions of the audience. Will they purchase the product? Did they change their opinion of the company? Will they vote for a candidate or for an issue?
Fairly unique to public relations research is the use of content analysis as a research method. Content analysis provides a way to objectively measure the messages of opinion leaders across the media. Contemporary public relations has gone beyond simply counting the number of press releases printed or minutes of client airtime (outputs); now the campaign’s key messages (Were they picked up? If so, how?) and the general tone of the messages (positive, negative, neutral) are evaluated as outcomes. Content analysis can be used on anything written or observable, such as corporate communications or media content. This can take the form of an analysis of the content or the tone or looking for a mention of certain things. This analysis can also include prominence (the location of the coverage), quality (includes tone, volume, prominence), and competitive analysis (looking at comparisons between coverage of competitors). Content analysis methodology can be conducted by human coders after extensive training or through the use of computer programs such as NVIV07, Diction, or The Ethnograph.
The topic of measurement has been controversial in public relations research. As the profession has moved from counting the products of a campaign, such as the number of clips or minutes of air time, to measuring the effects of a campaign, the question has become whether public relations can be measured. And if it can be measured, what exactly is measured and how should it be measured? Part of the measurement conundrum lies in the definition of what public relations is: Is public relations the free placement of messages, or is it more the management of relationships, reputation, trust, or credibility?
Outputs or Outcomes?
Since the first publicity campaign, success in public relations has centered on the producing of outputs. It was often measured by showing the client the “clipbook,” or copies of the output that had made it into print, on air, or in whatever mass medium was targeted. Those who buy into the definition that public relations is the free placement of messages in the media tend to refer to public relations’ success in terms of what it would have cost to place such messages in the media as a marketing strategy to provide estimates of influence as financial indicators. Hence, this point of view would measure success in terms of the cost of advertising in targeted media, or the “advertising value equivalency” (AVE) of the placement of the message. Another measure often used in this “measurement” is the “opportunity to see” (OTS) measure, which estimates message exposure.
Impressive piles of outputs were thought to have demonstrated public relations impact and probably did have some impact on a client’s perception of public relations success. Simple counting, however, cannot demonstrate an impact on the client’s public relations ROI. To demonstrate impact, the public relations campaign or program must demonstrate how its strategic use of those outputs has affected client final financial outcomes.
Complicating the matter are controversies related to what is an appropriate measure of public relations. There are two problems with AVE and OTS, and both are based on the way they are “calculated.” First, both are heavily influenced by circulation numbers in the print media. That is, an AVE’s monetary value is a function of circulation, which is a function of sales, and sales drive advertising costs. This can be a problem because circulation numbers are often exaggerated to drive up advertising costs. Second, circulation is often defined differently for street versus home sales, which are often multiplied by the average family size, insinuating that all members of a family will read or see the message. The same is true of broadcast media. Internet-produced messages have yet to yield a good way of measuring their impact, although “hits” and “key throughs” (going to a second page or site on the Internet page of question) have been examined and found lacking in validity or reliability.
Outtakes as Predictors of Outcomes
A different definition of public relations focuses public relations’ strategic place as a management tool. Stacks (2002), for instance, defines public relations in terms of the management of a client’s credibility. Others define public relations in terms of the management of client relationships and trust. Still others define public relations in terms of the management of client reputations. As such, public relations is seen as mediating the expectations that target audiences have of the client.
Stacks (2005) argues that a client’s credibility is a function of the public’s perceptions of trust in the client’s actions and products, its relationship with the client, and how it perceives the client’s historical reputation with the public. It is the role of public relations to manage these nonfinancial indicators to the client’s best interests. These then interact to influence the public’s perceptions of the client, which in turn affect the client’s ROI for public relations activities. He further argues that ROI comes from not only financial indicators such as sales, profits, and expenses but also from nonfinancial indicators and how they influence stock-and-stakeholder perceptions of the client.
Thus, from a management (as opposed to marketing) definition, public relations seeks to influence through the strategic production of key messages targeted at influentials (anyone who a target audience would see as a credible source of information), where the messages are factors that (1) set expectations and (2) help predict outcomes that establish public relations’ impact on client ROI. As noted above, the targeted audience in such instances would be editors or stock analysts, or anyone who might serve as an influencer (outtake) for a particular target audience or public, and measurement then would be whether the client’s key messages were communicated in the manner expected.
What is measured today in this approach is typically the content of influentials’ messages. The methodology employed is content analysis, a research method that provides both qualitative and quantitative interpretations of messages and is particularly important from a strategic management of the communication approach to public relations. Content analysis provides a measure of not only how many (counting) times a client’s name is found in the media but also how the client’s key messages have been received and communicated to the target media audiences. Thus, content analysis can provide several measures of public relations success across informational, motivational, and behavioral objectives and provide information for the refinement phase of the campaign.
The first measure establishes whether the client’s key messages were picked up and understood. If the messages were not being relayed to the larger target audiences, the influential was not serving a mediating purpose, and the public relations strategy must be rethought. Hence, the first measure provides feedback on whether public relations strategy is actually working. If the influential has picked up on the client’s key messages, how are those messages being internalized by the influential? Generally, content analysis today looks beyond the “clipbook” number of media mentions and tries to establish if the influential’s message about the client is positive, neutral, or negative. An excellent example of this is found in Michaelson and Griffin’s (2005) measurement of a MetLife media relations campaign.
Second, measures of public relations success can be established by comparing the client’s “share of voice” or the client’s share of messaging in the client’s particular industry against that of competitors (Jeffrey, Michaelson, & Stacks, 2006, 2007). Share of voice provides evidence of how the client’s key messages are being picked up and where they are being picked up, and then the share of voice is correlated with that of other competitors.
Third, measures of information/key message success can be assessed to see how the influentials are internalizing the messages and thus are motivated to extend the client’s credibility, trust, and reputation through relationships with the publics and relationships with the influentials. This yields a “third-party endorsement” strategy that provides extra credibility to the client’s messages (see Cameron, 1994; Grunig, 2000; Hallahan, 1999; Michaelson & Stacks, 2007).
Finally, the results of the measurement can be used to establish whether or not the public relations campaign has had an effect through surveys, focus groups, or in-depth interviews of selected individuals from the targeted publics in order to establish if the expected behavioral intent is strong enough to predict the outcomes as expected for the client’s business interests, thus either confirming public relations’ success or providing the feedback necessary to make changes in the campaign. Measurement, then, becomes an integral part of public relations practice.
Evaluation methods and tools differ greatly depending on whether the research is being conducted by an academic in the pursuit of extending knowledge or by the practitioner as a way of conducting a client campaign.
As noted earlier, scholars have criticized academic research as being too descriptive and not inferential; that is, academic public relations research has not defined or tested those mediating factors such as credibility, relationship, reputation, or trust enough to provide causal evidence of effectiveness. This is the result of two things: a lack of theory for an area as large as public relations and difficulty in measuring actual results in a very competitive practitioner world. It is also difficult to engage in theoretical research when the issue under study deals with crisis communications in today’s politically correct academic environment. Imagine if you will be trying to get, through a human subjects committee, a study that might cause confusion, concern, stress, or even embarrassment on the part of students (who make up most experimental study populations). Consider, too, whether a company would want its name or brand associated with a negative news report or negative rumor.
These problems leave most academic research in the realm of content analysis of public relations practice or descriptive methodologies employing in-depth interviews and focus groups or surveys. Some research (e.g., Carroll, 2006) has used industry data to model public relations outcomes, but getting access to data is often problematic and getting into the “real world” to conduct experimental research is both difficult and costly.
Michaelson and Stacks (2004, 2007), for instance, have conducted two studies seeking to establish whether a “multiplier effect” actually exists that provides public relations with an advantage over advertising. Practitioners have argued that public relations is anywhere between three and eight times more effective than advertising, but the “multiplier effect” has never been really tested. If the effect was obtained, they wanted to know how large the multiplier was. They questioned whether the industry-wide practice of arguing that public relations’ ability to inject a third-party endorsement would increase the client’s credibility, homophily (a variable that reflects how much individuals see a product brand as similar to themselves and their friends), product knowledge, and intent to purchase a product. They conducted two experiments. In the first, they randomly exposed undergraduate students to a public relations editorial endorsement of a product against a print advertisement, an Internet advertisement, and a radio advertisement and found no differences on any outcome. Furthermore, when conducting the study, they asked the students their media uses and then analyzed student use against that reported for the general public; they found no differences between student and general public media uses, thus ruling out that the students use of media could have influenced the results (and if there were differences, they should have come out in the Internet and print media exposures).
Intrigued by this finding, Michaelson and Stacks (2007) attempted a field experiment. A field experiment controls the content of the experimental stimuli but cannot control the location or distractions and must employ a convenience sample of research participants. They felt that the initial experiment was too complex—that is, it had too many different stimuli messages (although each student was exposed to one and only one message, the number of students per exposure was reduced due to the number of messages), so they focused only on the print editorial and print advertisement, exposing participants to onlythe editorial or advertisement in isolation from any other stories or advertisements. A total of 351 participants who read newspapers on a regular basis were recruited from six major malls across the United States. An equal number of participants (150 each) were randomly exposed to either the editorial or the advertisement; 50 participants were randomly assigned to a control group that received no experimental stimulus. Results found that exposure to either editorial or advertisement produced significantly higher ratings for product credibility, homophily, product knowledge, and intent to purchase the product. However, there were no differences between editorial and advertisement, except for higher homophily from participants exposed to the editorial. A third study that puts the editorial and advertisement on actual newspaper spreads is currently being conducted.
Although content analysis in practice typically looks at media coverage of an organization, person, or event, in academia, researchers often look at coverage of multiple organizations, people, or events or multiple messages about any one topic. DiStaso (2007) took this one step further when she conducted a content analysis looking at earnings releases from Fortune 500 companies compared with press coverage of earnings. She found that the local media paid greater attention to the positive aspects of the earnings releases (e.g., an increase in sales), while the national media focused on the negative aspects (the litigation and the decrease in net income).
Another example of academic research is Carroll’s (2006) study using structural equation modeling (SEM) to predict the influence of familiarity on corporate equity. By analyzing a sample of 1,500 participants, she found that familiarity does have a positive relationship with citizenship, reputation, and personality.
Evaluation methods and tools will differ from research phase to phase. In the developmental phase, the research may be based on extant secondary materials. Materials such as previous research conducted by the agency or corporation, industry publications, annual reports, and governmental publications may provide enough information to begin to plot public relations strategy for the problem at hand. Sometimes, secondary sources are not available, and primary data must be collected, typically in the form of in-depth interviews with influentials, focus groups of individuals from targeted audiences, or surveys of larger audiences or publics. In either case, the research conducted during the developmental phase should establish the baseline against which future research will be compared to measure success or failure during targeted dates in the refinement phase.
During the refinement phase, content analysis of key messages and surveys of intended audiences should be undertaken. Research at this phase establishes if the three major research objectives (information, motivational, behavioral) are being met as expected. Content analysis provides data on whether the client’s key messages are getting out and if they are understood (output to outtake). If not, then the campaign’s messaging strategy needs adjustment. Surveys and focus groups provide the data necessary to establish whether the information has caused the motivational outcomes anticipated in the campaign. Have attitudes or beliefs changed or been reinforced as required to meet the motivational objectives? If so, then, has the target audience through survey and focus group methodologies indicated an intent to behave as predicted?
At the evaluation stage, the public relations campaign’s results are then examined to see if they did indeed influence the business objectives. If the campaign was carefully researched and research was conducted throughout the refinement period, to include assessment of audience behavioral intentions, then the campaign should have a demonstrated impact on the client’s business outcome and ROI.
Understandably, the primary statistical tools used to analyze public relations research are simple descriptive statistics—frequencies, percentages, proportions, and now correlations (e.g., share-of-voice research). For the most part, the research does not use inferential statistics, but that trend is slowly changing as more statistically sophisticated academics and practitioners enter into public relations research. The movement from a stress on outputs to out-takes/outcomes provides both sets of researchers with the ability to use advanced statistical tools, and the movement toward measuring mediating variables has introduced into the research literature a number of highly sophisticated inferential tools.
The drive to measure “what we cannot see” (Stacks, 2002) in terms of attitudes and beliefs and concepts such as credibility, homophily, purchase intent, trust, and even operationalizing relationships has introduced factor analysis as a statistical tool in the academic literature and slowly into the practice (particularly through the literature available from the Institute for Public Relation’s Commission on Measurement and Evaluation, http://www.institituteforpr.org/about/measurement_commission). From this, academic research has moved from exploratory factor analysis, whereby the underlying structure of measuring instruments is determined (Stacks, 2002), to confirmatory factor analysis, where the historical instrument’s measurement structure is confirmed for the particular sample being observed (Stacks, in press).
The ability to create measures that are reliable and valid has led to the use of multivariate statistical tools such as multiple regression, path analysis, and sequential equation modeling (SEM). A number of academic researchers have used multiple regression as a tool to establish the effect of multiple mediator variables on public relations outcomes. Path analysis, which is best when used with “hard” data (e.g., economic, sales), takes the multiple regression and creates causal models of the direct and indirect impact of mediating variables on outcomes of interest (e.g., David, Kline, & Yang, 2005; Lee, 2005; Wan & Schell, 2007). Furthermore, path analysis has no limitation on the number of cases that it can analyze. SEM analysis, on the other hand, is best used with “soft” data (e.g., mediator variables such as attitude or belief measures) and begins by testing the measurement instruments’ reliability and validity and then produces causal models of the impact of direct and indirect mediating variables on the outcome of interest (e.g., Ki & Hon, 2007; Kim, 2001; Yang, 2007). Two other things differentiate path analysis from SEM. First, SEM is “theory driven,” that is the models created use theoretically derived relationships between the mediator and outcome variables, while path analysis is not. Second, SEM can only handle a limited number of cases, whereas path analysis can handle all the data that are available. Marketing definition-oriented researchers, then, should use path analysis as their modeling tool, while management, mediated-related definition-oriented researchers should use SEM as their modeling tool.
The statistical tools used in most daily public relations practice are simple descriptive statistics. This is due mainly to the lack of research sophistication in the average practitioner and an even greater lack of knowledge of statistical tools and their uses. Seldom will research be reported to a client that includes inferential statistics, and such research usually employs analyses on what has been labeled categorical data—data that are differentiated by categorization; variables such as sex (male/female), income (high/middle/low), and occupation (blue/white collar) are examples, as would be age or income when broken into categories. The statistical tools used in such research are not typically presented to the client, although they may be used when asked about confidence in the findings.
There are exceptions to the elementary use of statistical tools, and they are found mainly in the corporate communications side of public relations. Companies such as General Motors, Microsoft, and Texas Instruments use advanced statistical tools to model their data and are increasingly using the models to establish the impact of their practice on business outcomes and ROI in particular. This makes sense when considered from an internal communications perspective, where years of organizational communication and industrial psychology research have established management theory on what makes a successful organization. Interestingly, there are research firms that are taking the sophisticated modeling capabilities of path analysis and SEM and looking for causal indications of reputation and trust. The research basically lacks a good theoretical base and often attempts to put all possible variables into the model, hoping to find which variables offer direct causal paths to the outcomes of interest.
There are no data available that allow for a comparison of public relations research other than what has already been discussed. A study by the International Public Relations Association is examining the research needs of practitioners at various levels of experience, but for the most part, public relations research in the United States is far ahead of other country or region practices (Stacks, in press).
As noted above, public relations research is at a crossroads of sort. Interestingly, the Internet may provide the impetus for movement toward more sophisticated research methods and analytical tools (DiStaso, Messner, & Stacks, 2007). Before the Internet and the new social media, influentials were easily identifiable. Today’s influential is more often than not a blogger who has gained a following from what he or she has written. Before the Internet, an influential’s impact was felt on the circulation numbers associated with the media for which he or she wrote or aired in, and as we know, circulation numbers may not provide an accurate picture of influence or impact.
Today’s social media blogger influence can be directly assessed from how many people visit the blogger’s site, how many engage in communication with that blogger (indicating a relationship and credibility that builds quickly to a reputation and trust), and how long the blog-ger stays on the Internet. There are ways to assess the blog-ger’s number of communications, trace the blogger’s social network, and actually measure the density or capacity of the links between the blogger and his or her readers (Stacks & Watson, 2006). Social network analytical tools now provide ways of plotting, describing, and predicting blogger influence. Such data, once captured and analyzed, can then be used in other sophisticated modeling tools to establish predictive models of influence.
Public relations research encompasses all the traditional research methodologies and, increasingly, some of the more sophisticated analytical statistical tools. As the practice matures and settles on a direction, the practice of public relations research will become more sophisticated. Future research will still demonstrate the effect of public relations by simple counting of outputs, but more and more, it will move toward understanding and predicting the effects of mediating variables on business outcomes and establishing the effect of public relations on business ROI.