Self-talk, empowerment, and goal-directed behavior in Patrick & Hagtvedt

In “I don’t” versus “I can’t”: When empowered refusal motivates goal-directed behavior,” Patrick & Hagtvedt (2012) conduct four separate studies to test their research hypotheses related to the link between self-talk and goal-directed behavior. All four of the studies are quantitative in nature, meaning that they rely on numerical data and statistical analyses. Three of the four studies included in the Patrick & Hagtvedt (2012) research were quasi-experimental, and the fourth (Study 3: the field experiment) is non-experimental. Although the authors do not explicitly spell out the sampling methods they use, they suggest ways they may have ensured random selection and double-blindness. Both qualitative and quantitative research methods are essential to the social sciences. Qualitative research is used when the researcher is interested in the participants’’ subjective perceptions or opinions, and does not rely on numerical data (McLeod, 2017). Although no statistics or numerical data are used, qualitative research is still empirical in nature, as the researcher uses the scientific method to develop a hypothesis, uses systematic sampling and data assessment methods, and also interprets data according to the ethical ground rules of the social sciences. Qualitative research methods include case studies, observations, and interviews. The researcher also attempts to preserve the informant’s/participant’s own language to reduce the chance of researcher bias (McLeod, 2017). Moreover, qualitative research methods interpret results not in terms of numbers and statistics but in terms of themes that emerge from the data. For example, if a researcher conducted interviews, the transcripts of the interviews would be scanned and coded according to word or phrase usage and themes that the subjects addressed in their responses. Quantitative research by definition yields numerical and/or statistical outcomes. More technically objective than qualitative research, quantitative research seeks objective truths or facts about social phenomena (McLeod, 2017). Whereas qualitative research can lend valuable insight into the parts of human psychological and sociological realities that are subjective and therefore not quantifiable, quantitative research can demonstrate important trends and patterns yielding fixed and measurable data outcomes. Although some scientists may presume quantitative research to be more accurate and valid versus qualitative research because of the presumption of a “fixed and measured reality,” neither quantitative nor qualitative research necessarily captures the entire truth and neither approach is better or worse than the other in promoting scientific inquiry (McLeod, 2017, p. 1). As Newman (2016) puts it, “a true understanding of any phenomenon requires the use of both methods,” (p. 14). Research design plays a more important role in study validity and reliability than whether or not the study is quantitative or qualitative. However, quantitative research does seem impressive because of the ability to make statistical inferences. The most important benefit to using quantitative research methods is to determine causality between two variables, or at least to measure the direction and strength of a correlation between two or more variables. Measuring correlation and causation are not possible using qualitative methods. A mixed-methods approach may be warranted when studying certain psychological or social phenomenon. Because Patrick & Hagtvedt (2012) are interested in the behavioral outcomes of different self-talk patterns, a quantitative method is the more sensible approach. Behavioral outcomes are quantifiable. At the same time, self-talk is by its very nature a qualitative independent variable. It is difficult if not impossible to know what each participant is thinking; each person is in fact barely able to monitor his or her internal thoughts and beliefs systems. Still, the research questions related to the impact of different types of self-talk on goal-directed behavior are meaningful and have tremendous implications for various fields of research including health psychology, organizational psychology, and marketing. Experimental designs are the gold standard in the social sciences, because they offer the opportunity to measure the relationship between two or more variables. A true experiment is one that uses random assignment and which is also double-blind, meaning neither the researcher nor the participant is aware which participant is assigned to which group (Trochim, 2006). Moreover, a true experiment has both an experimental and a control condition. A quasi-experiment uses an experimental and control group but participants are not randomly assigned to each group. If there is no control group at all, then the study cannot be considered an experiment (Trochim, 2006). To measure the effects of either “I don’t” self talk versus “I can’t” self talk, the researchers assigned participants to one or the other experimental group. Without a control group that did not receive any self-talk directives, the studies cannot be considered true experiments. The field study has the most problematic sampling methodology because it relies on participant participation in a health and wellness seminar. However, all participants came from the same population pool and so it can be assumed that they started with the same baseline of goals related to the pursuit of specific personal healthcare goals. For Study 1, the researchers first assessed goal relevance for 120 undergraduate students and then presumably randomly assigned the participants to one of the two experimental groups, “I don’t” or “I can’t.” There were two different dependent variables including one self-assessment of perceived empowerment, measured on a Likert scale, and a more direct post-test in which each participant was asked to choose between two different snack options. In Study 2A, the sampling method was more robust, relying on online panel recruitment consisting of a population pool of as many as 3 million people. A total of 179 adults agreed to participate in the study, and as with Study 1, the researchers operationalized the goal-directed behavior to make sure all participants “had the same operative goal and the same temporal horizon” to achieve it (Patrick & Hagtvedt, 2012, p. 375). However, the method was to have participants “imagine” a scenario, which may not be a valid means of measuring actual motivation (Patrick & Hagtvedt, 2012, p. 375). Dependent variables were likewise subjective and had more to do with imaginary rather than real outcomes. In Study 2B, 120 undergraduates participated in a study that used both pretest and post-test. In this study, internal and external motivations were used as variables in addition to the self-talk condition. As with Study 2A, Study 2B did not measure actual behavioral change but only imaginary or projected change reported by the participants. Finally, in the third study, the researchers used a field experiment model using 30 women who registered with a health seminar as participants. Dependent variables were measured with goal persistence, but also by self-reports. Therefore, none of the four quasi-experimental studies has a large degree of internal or external validity even though the overall subject may prove relevant for future researchers seeking insight into how self-talk actually influences consumer behavior or health-seeking behaviors. Falling under the rubric of cognitive psychology, the Patrick & Hagtvedt (2012) research has design flaws that threaten internal and external validity. The research has real-world application and is therefore applied research, and yet the design flaws may render the study irrelevant. Using actual behavioral outcomes as dependent variables, plus designing the studies so that they were truly experimental instead of quasi-experimental or non-experimental, would also have made the Patrick & Hagtvedt (2012) research more robust.