Stuttering attitudes among Turkish family generations and neighbors from representative samples

Abstract

Purpose

Attitudes toward stuttering, measured by the Public Opinion Survey of Human Attributes-Stuttering (POSHA-S), are compared among (a) two different representative samples; (b) family generations (children, parents, and either grandparents or uncles and aunts) and neighbors; (c) children, parents, grandparents/adult relatives, and neighbors of the same family/neighbor units vs. individuals from different family/neighbor units; and (d) attitudes from one Turkish city with an international database archive.

Methods

Following a school-based, three-stage, cluster probability sampling scheme, two sets of children, parents, grandparents/adult relatives, and neighbors (50 each) in Eskişehir, Turkey (PROB1 and PROB2) completed Turkish translations of the POSHA-S. The POSHA-S measures attitudes toward stuttering within the context of other attributes, such as obesity and mental illness.

Results

Both replicates of the sampling procedure yielded strikingly similar attitudes for stuttering between children, parents, grandparents/adult relatives, and neighbors in PROB1 vs. PROB2, and between all pair-wise comparisons within PROB1 and PROB2. By contrast, attitudes toward obesity and mental illness were dissimilar. Correlations were small to moderate among attitudes of the same family/neighbor units but were essentially nonexistent between different family/neighbor units. Attitudes toward stuttering in Eskişehir were estimated to be less positive than attitudes from a wide range of samples around the world, although exceptions occurred.

Conclusions

A school-based probability sampling procedure yielded consistent findings that are likely different from results from convenience samples. Families appear to be an important influence in determining public attitudes toward stuttering and other human attributes.

Educational objectives: The reader will be able to: (i) identify similarities and differences among attitudes toward stuttering across generations; (ii) identify similarities and differences among attitudes toward stuttering in Turkey vs. other places in the world; (iii) describe a school-based probability sampling scheme; (iv) describe advantages of using a standard instrument to measure public attitudes toward stuttering.

Introduction

Since the 1970s, popularized by seminal studies of Dean Williams and his colleagues (Woods and Williams, 1976, Yairi and Williams, 1970), numerous studies have shown that samples of nonstuttering adults hold perceptions and beliefs about the approximately one percent of people who stutter that have been interpreted to reflect bias, misinformation, stereotypes, and/or stigma (e.g., Craig et al., 2009, Cooper and Cooper, 1996, Crowe and Walton, 1981, Doody et al., 1993, Dorsey and Guenther, 2000, Gabel, 2006, Gabel et al., 2004, Healey et al., 2007, Hulit and Wirtz, 1994, Hughes et al., 2010, Klein and Hood, 2004, Panico et al., 2005, Ruscello et al., 1988). A similar large number of studies have explored the attitudes about – or reactions to – stuttering among children (e.g., Blood et al., 2003, Evans et al., 2008, Ezrati-Vinacour et al., 2001, Griffin and Leahy, 2007, Horsley and FitzGibbon, 1987, Hartford and Leahy, 2007, Langevin, 2009, Langevin et al., 1998). The unmistakable conclusion from these investigations is that those who stutter live in a social environment where stuttering is viewed negatively. Moreover, the attitudes have further been regarded as important factors underlying unacceptable behavior ranging from teasing and bullying to job discrimination.

What are the principal reasons for the negativity associated with stuttering by the nonstuttering majority? Some (e.g., Klassen, 2002, Langevin et al., 2009) have suggested that lack of association with people who stutter may be partly responsible, reporting that attitudes were less negative when respondents rated known individuals who stutter rather than hypothetical or all individuals. Importantly, this finding has not been consistently replicated (e.g., Gabel et al., 2004). White and Collins (1984) advanced the notion that most people have “stuttered” at one time or another in their lives, typically when they were stressed, anxious, excited, or embarrassed and thus attribute the same emotions or characteristics to people with chronic stutters. Supporting but extending White and Collins's view, respondents in MacKinnon, Hall, and MacIntyre's (2007) study rated a hypothetical male with temporary stuttering more negatively than a hypothetical male chronic stutterer for the very attributes that characterize the stuttering stereotype, i.e., the person with the temporary stutter was judged to be more afraid, fearful, nervous, tense, anxious, introverted, and unpleasant than the person with chronic stuttering. The respondents were said to have made a quick but incomplete “anchoring-adjustment” decision wherein they first thought about their own instances of temporary disfluency that probably occurred during periods of stress. Accordingly, they attributed strong emotions to the temporary stuttering individual. But, suspecting that the chronic individual would be different, they incorrectly assumed that he had adapted to his emotions; hence, they rated him more positively. Boyle, Blood, & Blood (2009), citing research on stigma associated with mental illness, postulated that presumed psychological etiology fosters negative attitudes and showed that highlighting a psychological cause or unknown cause of stuttering fosters more negative attitudes than highlighting a genetic cause.

It is worth noting that Wendell Johnson, whose views greatly influenced those of Dean Williams, argued vociferously that family values promoting (or expecting) higher speech standards or demands, were responsible for the social environment of the stuttering speaker, and in fact, caused the problem in the first place (e.g., Johnson, 1961). Whereas little evidence has supported Johnson's “diagnosogenic theory” of stuttering (e.g., Van Riper, 1971), Johnson's idea that parents are instrumental in passing various attitudes to their children, aside from “causing” stuttering, has not been challenged. Indeed, it is widely believed that many of our deeply held attitudes and values, and even stuttering stuttering attitudes, are passed on from generation-to-generation (e.g., Bengston et al., 2002, Betz et al., 2007).

Johnson also believed that stuttering was highly culturally limited, basing his argument on linguistic determinism, i.e., that the distinctions in one's language determine a great deal of the content of one's thoughts and actions (e.g., Whorf, 1956). He wrote that the American Indians had no word for “stuttering” and, hence, had little or no stuttering. This view also turned out to be erroneous (Zimmermann, Liljeblad, Frank, & Cleeland, 1983), but highlights a recurring notion that stuttering should be studied cross-culturally. A number of investigators have shown that some of the negative attitudes reported in studies from North America, Western Europe, and Australia have been replicated elsewhere in the world, but with some differences (e.g., de Britto Pereira et al., 2008, St. Louis et al., 2005, St. Louis and Roberts, 2010, Xing Ming et al., 2001).

Arguably, one of the principal obstacles to obtaining a body of replicated evidence that different age groups hold similar or different attitudes toward stuttering, or that these attitudes are similar or different interculturally, has been the absence of standard measures of attitudes specifically designed for intergenerational and intercultural comparisons. A related obstacle has been the reliance on convenience sampling as opposed to representative sampling. Only a few reported studies of attitudes toward stuttering have attempted to use representative samples (e.g., Al-Khaledi et al., 2009, Craig et al., 2003, de Britto Pereira et al., 2008, Evans et al., 2008, Özdemir et al., in press, Van Borsel et al., 1999, Xing Ming et al., 2001).

For more than a decade, the Public Opinion Survey of Human Attributes-Stuttering (POSHA-S) has been under development as a standard measure of stuttering attitudes (St. Louis, 2005, St. Louis, in press). The POSHA-S is the first instrument to be developed in the International Project on Attitudes Toward Human Attributes (IPATHA). Its two overarching objectives are to provide a standard instrument that can be used to measure attitudes among literate populations worldwide and that can measure the effects of initiatives designed to improve public opinion about stuttering. Users of experimental versions of the POSHA-S have been asked to contribute their respondents’ ratings to a growing database archive that can be used to determine what attitudes are comparatively more positive, negative, or similar to other samples. Obviously, these comparison “norms” will not be static, but will change as more and different populations are sampled. For example, St. Louis and George (2008) reported on two groups that, to date, have documented the most positive ratings, i.e., leaders in the stuttering self-help movement and Board Recognized Specialists in Fluency Disorders. It is reasonable to assume that the precision of comparisons will increase as (a) the number of samples and respondents continue to grow (St. Louis, in press) and (b) data from representative samples are added to the archive.

The POSHA-S has been described extensively elsewhere (e.g., St. Louis, 2011, St. Louis, in press, St. Louis et al., 2008), but the following aspects are germane to this study. The POSHA-S is a printed survey instrument, whose design was informed by epidemiologic principles (St. Louis et al., 2008), and requires about 10 min to complete. Circa January, 2011, it has undergone field tests in 12 countries in eight different languages. The instrument requests demographic information, general comparisons of stuttering with four other “anchor” attributes (obesity, mental illness, intelligence, and left handedness), and detailed beliefs about – and self-perceived reactions to – stuttering. The rationale for including attributes other than stuttering in the POSHA-S was to be able to consider attitudes toward stuttering within the context of other negative, neutral, or positive human conditions (St. Louis et al., 2008). Most reports to date have focused on documentation of psychometric and other characteristics of the POSHA-S: test–retest reliability (St. Louis, Lubker, Yaruss, & Aliveto, 2009), construct and concurrent validity (St. Louis, Reichel, Yaruss, & Lubker, 2009), internal consistency (Al-Khaledi et al., 2009), translatability to other languages (Al-Khaledi et al., 2009, St. Louis, 2005, St. Louis and Roberts, 2010), comparison with a growing archive of POSHA-S data (St. Louis, in press), probability vs. convenience sampling (Özdemir et al., in press), and paper-and-pencil vs. online administration (St. Louis, Hancock, & Remley, 2010). The instrument has also been shown to be sensitive to changes in attitudes produced by coursework in fluency disorders in SLP graduate students (St. Louis, Reichel, et al., 2009) and either a live oral presentation by a person with moderate-severe stuttering or a professionally prepared video on stuttering in high school students (Flynn & St. Louis, 2011).

These and other studies have focused on specific attitudes as well (e.g., Al-Khaledi et al., 2009, St. Louis et al., 2005, 2011; St. Louis & Roberts, 2011). Ratings and demographic information from 3257 respondents from 91 comparisons (circa January, 2011) have been consolidated into the POSHA-S archive or database. Importantly, all but three of these sample comparisons were carried out using convenience sampling procedures (St. Louis, 2011). Based on a recent report, it is likely that convenience samples yield different attitudes than probability representative samples (Özdemir et al., in press).