Diego Vargas Barona
The Graduate Center, CUNY
 
 
Introduction

Language attitudes are central in the interaction between speakers, especially between native speakers of English (NSs) and non-native speakers of English (NNSs).  Many sociolinguists have concentrated on the study of the attitudes that speakers have toward accents and they have discovered different aspects of this relationship. Many studies about language attitudes have focused their attention on finding out the attitudes of NSs toward accented speech, but only a handful of studies have focused their attention on the attitudes of NNSs towards non-native accents. Nor have the attitudes of NSs and NNSs been compared to examine any similarities and differences that might exist between the two groups. Language attitudes play a crucial role in interpersonal communication, so it is essential to study the attitudes that NSs and NNSs exhibit toward non-native accents. In addition, studies like the present one may be very important because thousands of non-native speakers communicate with native speakers in social and economic settings that are critical to the welfare of many people. The results of this study will help make English as second language (ESL) instructors, ESL students, and the population of native speakers more aware of how people perceive non-native accents. The following review reports on studies that have produced results that are worth mention.
 
1. Previous Studies

The outcomes of some studies have shown that NSs possess negative attitudes toward regional varieties of English. In a study conducted in the United Kingdom, Giles (1971) showed that speakers using “standard” pronunciation were rated more positively than speakers using a regional accent; the former were rated as more competent, self-confident, and better educated. A subsequent study conducted by Cargile and Giles (1998) reveals that a complicated relationship can exist between attitudes toward regional accents and non-native accents. The data collected in this study demonstrated that NSs of British English rated moderately Japanese-accented speech as high as English dialects in the United States in terms of status-related traits; however, NSs rated the Japanese-accented speech lower on traits related to social attractiveness.

Other studies have collected data that show that sometimes NSs possess negative attitudes toward non-standard varieties of English (Brennan and Brennan, 1981). In this study, it was found that both Mexican-American and Anglo-American listeners rated standard American English speech more favorably than Mexican-American accented speech in terms of social attractiveness and solidarity. Similarly, Kamisli and Dugan (1997) performed a study in which university students from Turkey were asked to make judgments about three different American English dialects (Eastern, African-American, and Southern accents). Speakers of these dialects were evaluated in terms of their personalities and their suitability for employment as a teacher of English in Turkey. In terms of personality traits, the Eastern accent and the Southern accent were rated in a similar way and the African-American accent was rated lowest. However, in the open-ended questions, NNS evaluators stated that the Eastern accent was more suitable for employment than Southern and African-American accents.

Negative judgments of accented speech are not limited to native varieties of English. Johnson and Jenks (1994), Podberesky, et al. (1990) and Giles, et al. (1995) conducted studies that produced results in which NSs judged non-native accents more negatively than ‘standard’ speech in terms of social status, intelligence, and education. In another study, Lindemann (2003) focused on NSs’ reactions to Korean-accented speech but did not disclose the ethnicity of the speakers to the evaluators. In this study, the researcher added an additional component and instructed the NS evaluators to try to figure out where the speaker was from, and then rate him/her using a scalar-rating sheet. NSs identified Korean-accented speech as either “Indian,” “Latino,” “Japanese,” “Korean,” “Chinese,” or “Asian,” although the researcher did not supply these categories.  It was interesting to see that the speakers identified as “Asian” were rated more positively in terms of status and education than the speakers identified as “Indian,” “Latino,” “Japanese,” “Korean,” or “Chinese.”  Furthermore, Bresnahan, et al. (2002) performed a study that evaluated attitudinal (e.g. attractiveness, status, dynamism) and affective (e.g. pleasantness, arousal, dominance) responses toward non-native accents based on variations in role identity and intelligibility. The evaluators were NS university students from the United States; they were asked to listen to and evaluate several recorded speeches of NSs and NNSs. The evaluators were instructed to respond to a questionnaire that included four items that measured comprehension of the recorded speech. Speech samples of American English were rated more favorably than non-native accents, and intelligible non-native accented speech was rated more positively than unintelligible non-native accented speech. Additionally, not all non-native speakers are stigmatized in every category; a study conducted by Cargile (1997) evaluated the attitudes of Anglo-Americans and Asian-Americans towards Chinese-accented speech. The results obtained in this study showed that NSs evaluated Chinese-accented speech similarly to standard American English speech in terms of status-related traits, but they were rated lower on attractiveness.

While many studies have attempted to measure the attitudes that NSs have toward accents, there appear to be few studies that deal with NNSs’ attitudes toward non-native accents. Chiba, et al. (1995) conducted a study in Japan in which they gathered data that showed that NNSs have negative attitudes towards non-native accents. The researchers selected Japanese students of English as evaluators for this study. They instructed them to rate the personalities of NSs and NNSs by listening to their speech. The speakers under evaluation were NSs from the United States and England and the NNSs were from Japan and other Asian countries. In this study, the NNSs rated native accents more positively than non-native accents in terms of social status, intelligence, and educational background.  The authors of this study point out that the Japanese raters’ preference for American and British accents may be due to the fact that these accents are presented as ideal models in English as a foreign language schools in Japan. It is worth pointing out that the previous studies on NNSs’ attitudes towards non-native accents were done in an English as a Foreign Language (EFL) environment. It remains to be demonstrated that these results will be similar in an English as a Second Language (ESL) environment.

As shown in the previous literature review, we can see that the majority of studies have concentrated on finding out the attitudes of NSs toward accented speech, but very few have concentrated on the attitudes of NNSs towards non-native accents. Nor have the attitudes of NSs and NNSs been compared to examine any similarities and differences that might exist between the two groups. These issues are addressed in the present study.
 
2. The Study

Language attitudes play a significant role in the communication process between NSs and NNSs and both NSs and NNSs possess negative attitudes towards non-native accents as shown in the literature review above. Because communication among all members of society is important, it is vital to study language attitudes toward groups that are becoming part of American society so that they can be recognized and, perhaps, addressed. According to the 2000 U.S. Census, people of Spanish, Arabic, and Korean language backgrounds are some of the largest minority groups in New Jersey. It is important, therefore, to study the attitudes that speakers have towards the accents characteristic of native speakers of these languages. The present study focuses on these three groups of speakers and on the evaluations of their accents.

In addition to examining the relative ratings of accented speech as opposed to unaccented speech, the attitudes indicated by NSs and NNSs were compared to discover any similarities and differences that may exist between these groups. This comparison of NS vs. NNS attitudes toward non-native accents has yet to be explored.
 
Research Questions

In order to contribute to an improved understanding of language attitudes, the present study is designed to answer the following research questions:

Research question 1: Will non-native accents be rated more negatively than native     accents?

Research question 2: Will native speakers evaluate accents differently from non-native speakers?

Research question 3:  Will Korean, Spanish, and Arabic accents be rated differently?
 
3. Methodology and Materials

The speech of three male NNSs, representing three different language backgrounds, Korean, Spanish and Arabic, and three male NSs was recorded. The NNSs form part of their respective communities in Hackensack, Paterson, and Ridgefield, New Jersey.  The readers were chosen on the basis of their answers to eight demographic questions to make sure they had a similar background. (Appendix A).  The NS readers are New Jersey residents between the ages of 25 and 45. The NNSs have understandable accents that are similar in degree as perceived by the investigator. Each reader was asked to read a paragraph aloud, resulting in a reading of approximately 45 seconds. The paragraph was of neutral content and was obtained from the Slate website. (Appendix B). Their speech was recorded using a digital audiotape recorder.
 
3.1 Participants

Evaluators were 15 female NSs and 15 female NNSs who listened to and evaluated the speech of the native and non-native readers. The evaluators were people from the general public that live and work in Northern New Jersey. The language backgrounds of the NNS evaluators were: two Chinese, one Korean, three Portuguese and nine Spanish. They have been living in the United States for more than three years and learned English in an academic setting. Evaluators were asked a set of demographic questions to make sure they had a similar background before they were selected for the study. (Appendix C).  Some evaluators preferred not to answer some questions on the questionnaire; minimal personal information from the evaluators was required so as not to discourage their participation in this study.
 
3.2 Procedure

The investigator explained to the evaluators that they would listen to several speakers read a paragraph and would be asked to give their first impression of each speaker’s personality by filling out a rating sheet adapted from one developed by Osgood, et al. (1957) consisting of a list of descriptors. The rating sheet was adapted by omitting one of the descriptors, nervousness, which is not relevant to this study.  (Appendix D)

The rating sheet allowed the evaluators to rate each speech sample on a scale between one and six. Higher marks indicated more positive evaluations and vice versa. After each reader finished reading the paragraph, there was a pause of one to two minutes in which the evaluators scored the accent using the rating sheet described above.
 
4. Data Analysis

The ratings given for each of the 19 descriptors by each of the 30 evaluators for each of the six readers, a total of 3,420 data points, were examined. This number was reduced by grouping the descriptors into three categories, Competence, Integrity and Social Attractiveness, and computing the means for each category for each speaker for each evaluator. The first category, called Competence, was based on seven descriptors: ambition, intelligence, self-confidence, leadership, prestige, stability, and wealth. The scores for each descriptor for each reader were added and the result was divided by seven to obtain the mean score for this category. The second category, called Integrity, was based on five descriptors: honesty, good heartedness, good character, generosity, and reliability. The scores for each descriptor for each reader were added and the result was divided by five to obtain the mean score for this category. The third category, called Social Attractiveness, was based on seven descriptors: sense of humor, friendliness, likeability, entertainingness, cleanliness, height, and good looks. The scores for each descriptor for each reader were added and the result was divided by seven to obtain the mean score for this category. Thus, there were mean scores for each of the three categories for each participant for each reader, for a total of 540 means.

In order to discover if evaluators rate non-native accents more negatively than native accents, the mean of the scores given to native readers for each category and the mean of the scores given to the non-native readers for each category were computed and plotted on a chart.

The results show that non-native accents were rated more negatively than native accents.  In the category of Competence, NS readers scored 4.78, which is 1.43 points higher than 3.35, the score given to NNS readers (Table 1 and Figure 1).  In terms of Integrity, NS readers scored 4.58, which is .58 points higher than 4.0, the score given to NNS readers. (Table 1 and Figure 1).  In the category of Social Attractiveness NS readers scored 4.9, which is 1.35 points higher than 3.55, the score given to NNS readers (Table 1 and Figure 1).

 

Table 1. Overall Ratings for NSs vs. NNSs

Figure 1. Comparison of Means for NS and NNS Readers

 

In order to learn if NSs evaluate accents differently from NNSs, the means of the scores on each of the three categories given for NS readers and NNS readers by NS evaluators were looked at and compared to the means of the scores on each of the three categories given by NNS evaluators. The results were plotted on a chart to represent the information graphically. (Figures 2 and 3).

The results show that NSs evaluated native accents differently from NNS evaluators.  NS evaluators gave NS readers a rating of 5.1 in terms ofCompetence. This rating is .64 points higher than 4.46, the rating given by NNS evaluators to NS readers for this category. NS evaluators gave NS readers a rating of 4.5 to NS readers in terms of Integrity. This rating is .3 points lower than 4.8, the rating given by NNS evaluators to NS readers for Integrity.  NS evaluators gave a rating of 5.1 to NS readers in terms of Social Attractiveness. This rating is .4 points higher than 4.7, the rating given by NNS evaluators to NS readers for this category. (Table 2 and Figure 2).

 

Table 2. NS and NNS Evaluators’ ratings for NS readers

Figure 2. Comparison of means given by NS and NNS evaluators to NS readers

The results show that NSs evaluate non-native accents differently from NNS evaluators.  NS evaluators gave a rating of 3.3 to NNS readers in terms ofCompetence. This rating is .5 points lower than 3.8, the rating given by NNS evaluators to NNS readers for this category. NS evaluators gave a rating of 3.8 to NNS readers for Integrity. This rating is .4 points lower than 4.2, the rating given by NNS evaluators to NNS readers in terms of Integrity.  NS evaluators gave a rating of 3.6 to NNS readers in terms of Social Attractiveness. This rating is .1 point higher than 3.5, the rating given by NNS evaluators to NNS readers for Social Attractiveness. (Table 3 and Figure 3).

 

Table 3. NS and NNS Evaluators’ ratings for NNS readers

Figure 3. Comparison of means given by NS and NNS evaluators to NNS readers

In order to learn if Korean, Spanish, and Arabic accents are rated differently, the mean scores for the Competence, Integrity, and Social Attractivenesscategory for each NNS reader were compared.  The results were plotted on a graph to illustrate the relationship.

The results show us that the Korean-accented speech was rated higher in terms of Competence 4.1, which is .5 points higher than 3.6, the score given to the Spanish-accented speech. Additionally, the Korean-accented speech was rated 1.2 points higher in terms of Competence than the Arabic-accented speech that received a 2.9 score (Table 4 and Figure 4).

Table 4. Mean scores for the Competence category

 

Figure 4. Comparison of mean scores for the Competence category

 

In terms of Integrity, the Korean-accented speech and the Spanish-accented speech obtained similar ratings, 4.5 and 4.4 respectively.  The Arabic-accented speech scored 3.1, which is 1.4 points lower than 4.5, the score given to the Korean-accented speech, and 1.3 points lower than 4.4, the score given to the Spanish-accented speech. (Table 5 and Figure 5).

 

Table 5. Mean scores for the Integrity category

 

Figure 5. Comparison of mean scores for the Integrity category

The results show us that the Spanish-accented speech was rated higher in terms of Social Attractiveness 4.0, which is .5 points higher than 3.5, the score received by the Korean-accented speech, and it is 1.0 points higher than 3.0, the score received by the Arabic-accented speech. (Table 6 and Figure 6)

Table 6. Mean scores for the Social Attractiveness category

Figure 6. Comparison of mean scores for the Social Attractiveness category

 
5. Discussion

This study sought to answer three questions: Will non-native accents be rated more negatively than native accents? Will native speakers evaluate accents differently from non-native speakers? Will Korean, Spanish, and Arabic accents be rated differently?

To answer the first question, are native accents rated differently, the results show that NS and NNS evaluators rated non-native accents more negatively than native accents in every category. On the basis of previous studies such as Johnson and Jenks (1994), Podberesky, et al. (1990), Giles, et al. (1995) and Lindemann (2003), these results were not unexpected.

The second question asked whether the evaluations of NSs and NNSs would differ. A glance at Tables 2 and 3 and Figures 2 and 3 reveals that NS and NNS evaluate accents differently in terms of Competence, Integrity and Social Attractiveness. NS evaluators gave NS readers higher scores on Competence andSocial Attractiveness as compared to the scores given by NNS evaluators to NS readers; however, NSs gave NS speech lower scores on Integrity than NNS evaluators did. On the other hand, NS evaluators gave NNS readers lower ratings as compared to the ratings given by NNS evaluators to NNS readers in terms ofCompetence and Integrity (D= .5 and .4 respectively) but NS and NNS evaluators gave similar ratings to NNS readers in terms of Social Attractiveness.  It was interesting to see that, though NNSs rated non-native accents more negatively than native accents, the ratings given by NNS evaluators toward non-native accents were higher than the ratings given by NS evaluators toward non-native accents. This may be because some NNSs may associate more positively with some of the NNSs in this study and may have had positive experiences with other NNSs.

The results also confirm the findings of Chiba et al. (1995) that NNSs also posses negative attitudes toward non-native accents. This may tell us that even though NNS evaluators may interact more with NNSs in their own communities, apparently they react more positively to native accents and attribute more desirable qualities to NSs.  Preference for a native accent by NNSs may be due to the fact that a native accent may be considered by NNS evaluators as the ideal model to follow in English pronunciation.

The third research question asked whether or not there would be differences in evaluation among the three different non-native accents. The results of this study tell us that Korean, Spanish, and Arabic accents were rated differently even though the evaluators were never told which accent they were listening to and it is unknown whether NSs or NNSs distinguished among the non-native speech samples presented in this study.  The Korean-accented speech was rated higher than the Spanish-accented speech and the Arabic-accented speech in terms of Competence. The Korean and the Spanish accents were rated similarly and higher than the Arabic accent in terms of Integrity. The Spanish-accented speech was rated higher than Korean and Arabic accents in terms of Social Attractiveness. We can also notice that the Arabic-accented speech was rated lower than other accents in every category. If the evaluators recognized that the accent was Arabic, this may be connected to the changes in general public opinion towards Arabs due to events such as the 9/11 terrorist acts and the current wars in the Middle-East. In addition, it is known that the Arab communities tend to live in areas where the population is mostly of Arab descent and the number of immigrants from this background is much lower than the number of Korean and Spanish immigrants. Similarly, evaluators may have rated Spanish accents higher than the Arabic accent and equal to the Koreans because Spanish and Korean accents have become somewhat “acceptable” in northern New Jersey, where the foreign-born Hispanic and Korean population is increasing rapidly. In addition, the fact that the majority of NNSs evaluators’ first language was Spanish might have influenced the ratings given to the readers of similar backgrounds.
 
6. Conclusion

The present study investigated whether a non-native accent plays a significant role in influencing the listener’s judgment about the speaker’s Competence,Integrity and Social Attractiveness. The results obtained show that NS and NNS evaluators gave higher ratings to NS readers than those obtained by NNS readers. In addition, the present study shows that NNSs also have negative attitudes toward non-native accents, and the three non-native accents are evaluated differently. We learned that the Arabic accent was rated much lower than accents from Korean and Spanish language backgrounds, though there is no evidence that tells us if the evaluators were able to distinguish the language backgrounds in the speech samples. This new information can be used to improve the relations between native and non-native speakers because they can understand better what type of attitudes can be attributed to accent alone.

The results of this study showed that accent has an important impact on the evaluation of a speaker’s personal characteristics and that language may affect whether or not a non-native speaker is perceived as someone with positive or negative qualities. In other words, the perception of accents plays a great role in the judgments of others and can influence how and why a person has positive or negative feelings towards someone that speaks with a non-native accent.
Many more questions about attitudes toward accented speech remain to be explored. The findings of this study should be interpreted within its limitations. The nature of this study was exploratory and, so, descriptive statistics were appropriate to analyze the results. Exploratory studies pave the way for future work in which hypotheses can be tested using inferential statistics.

There are issues that may need to be considered for further research, for example: gender; speech samples of female voices can be used to determined if NSs and NNSs’ reactions are similar for each language background when gender changes. In addition, speech samples from more language backgrounds could be investigated to discover if NSs and NNSs produce similar ratings. Furthermore, what about the first language of the NNS evaluator? For example, would Arabic-speaking evaluators give Arabic speakers higher scores? We could use the same number of evaluators from the same language backgrounds as those that were rated in this study to find out if we obtain similar ratings. We can also use evaluators that come from language backgrounds that are different from the readers’ to avoid any cultural similarities between evaluators and readers. Our next step could be to use these results to state a hypothesis and use statistical tests to measure the difference in group means; this will allow us to learn if the results are statistically important.

According to the 2000 U.S. Census, the immigrant population is increasing rapidly. Because thousands of non-native speakers communicate with native speakers in settings that are critical to the well-being of many people, studies like the present one are very important. Furthermore, the issue of how to assist the immigrant population in learning English and improving their pronunciation skills must be addressed. A significant contribution of this study concerns raising the awareness of ESL instructors, ESL students, and the population of native speakers, about the perceptions of non-native accents.

In conclusion, it is known that communication among all members of society is important; so, it is vital to study language attitudes toward groups that are becoming part of American society so that they can be recognized and, perhaps, addressed. Society will benefit greatly if this is done.
 
 
Bibliography

Brennan E., & Brennan S. (1981). Accent scaling: language attitudes and reactions toward Mexican-American English speech. Language and Speech, 24(3), 207- 221.

Bresnahan, M. J., Ohashi, R., Nebashi, R., Liu, W. Y., & Shearman, S. (2002). Attitudinal and affective response toward accented English. Language and Communication, 22, 171-185.

Cargile, A.C. (1997). Attitudes toward Chinese-accented speech: An investigation in two contexts. Journal of Language and Social Psychology, 16(4), 434-443.

Cargile, A. C., & Giles, H. (1998). Language attitudes toward varieties of English: An American-Japanese context. Journal of Applied Communication Research, 26(3), 338-356.

Chiba, R., Matsuura, H., & Yamamoto, A. (1995). Japanese attitudes towards English accents. World Englishes, 14, 77-86.

Giles, H. (1971). Patterns of Evaluation in reactions to R.P., South Welsh and Somerset accented speech. British Journal of Social and Clinical Psychology, 10, 280–281.

Giles, H., Williams, A., Mackie, D., & Rosselli, F. (1995). Reactions to Anglo and Hispanic-American accented speakers: Affect, identity, persuasion, and the English-only controversy. Language and Communication. 15(2), 107-120.

Johnson, R., & Jenks, F. (1994). Native speakers’ perception of non-native speakers: related to phonetic errors and spoken grammatical errors. Baltimore, MD: Paper presented at the annual meeting of the Teachers of English to Speakers of Other Languages. (ERIC Document Reproduction Service No. ED372653)

Kamisli, S., & Dugan, S. (1997). Non-native speakers´ speech perception of native speakers. (ERIC Document Reproduction Service No. ED461266)

Lindemann, S. (2003). Koreans, Chinese or Indians? Attitudes and ideologies about non- native English speakers in the United States. Journal of Sociolinguistics, 7(3), 348-364.

Mallaby, S. (2006). The 60 largest American charitable contributions of 2005. Retrieved February 21, 2006, from http://www.slate.com/id/2136384/

Osgood, C., Suci, G., & Tannenbaum, P. (1957). The measurement of meaning. Urbana, Ill.: University of Illinois Press.

Podberesky, R., Daluty, R., & Feldstein, S. (1990). Evaluations of Spanish and Oriental- accented English speakers. Social Behavior and Personality, 18, 53-63.

United States Census Bureau, (n.d.). Retrieved March 15, 2006, from http://www.census.gov/prod/2003pubs/c2kbr-34.pdf
 
 
Apendix
 
APPENDIX A: Reader #:__________ Tape #:____________ Please answer the following questions:

1. Where were you born? ______________________________________________

2. Native Language ___________________________________________________

3. Is your age between 25 and 45? ________________________________________

4. What do you do for a living? __________________________________________

Answer the following questions if English is not your native language.

5. Other languages ___________________________________________________

6. How old were you when you first began to study English? __________________

7. How did you learn English? (Academically or naturalistically) _______________

8. How long have you lived in an English-speaking country? Which country? ____________________________________________________________________
APPENDIX B:

To the reader: Please silently look at the following paragraph for a minute. If there are words that you do not understand, please ask the investigator.

Ted Turner always got a buzz from giving. I had breakfast with him once in a sedate Washington hotel, and he explained how charity could juice him. He’d made his first big donation back in the 1980s, and his hand had trembled uncontrollably as he signed away his chances of becoming the world’s richest man. But having given money away once, he found he wanted to give more; charity became compulsive. As Turner explained this, he yelled and waved his arms around for emphasis, alarming the sleepy breakfasters at nearby tables. Giving could be an addiction, like alcohol or drugs, Turner proclaimed. “Like sex!” he roared enthusiastically.

Taken from http://www.slate.com/id/2136384/ 2/21/06
APPENDIX C: Evaluator #:__________ Please answer the following questions:

1. Where were you born? ______________________________________________

2. Are you between the ages of 18 and 25? _________________________________

3. What is your major? ________________________________________________

Answer the following questions if English is not your native language.

5. What is your native language? ________________________________________

6. How old were you when you first began to study English? __________________

7. How did you learn English? (Academically or naturalistically) _______________

8. How long have you lived in an English-speaking country?___________________

9. Which country? ____________________________________________________
APPENDIX D: Evaluator #:__________ Tape#:__________ Native language______________

In this experiment you will rate each voice on the characteristics listed below. Listen to each voice and begin to score as soon as you feel you can judge what type of person each speaker is. Give a rating for each trait. Do this in the same way as you do when listening to someone’s voice for the first time on the telephone or radio.

Good looks                          very little                             1  2  3  4  5  6                            very much

Leadership                           very little                             1  2  3  4  5  6                            very much

Sense of humor                    very little                              1  2  3  4  5  6                           very much

Intelligence                           very little                              1  2  3  4  5  6                           very much

Honesty                               very little                              1  2  3  4  5  6                           very much

Self-confidence                    very little                              1  2  3  4  5  6                           very much

Height                                  very little                              1  2  3  4  5  6                           very much

Friendliness                          very little                              1  2  3  4  5  6                           very much

Generosity                            very little                              1  2  3  4  5  6                           very much

Entertainingness                    very little                              1  2  3  4  5  6                           very much

Good-heartedness                very little                              1  2  3  4  5  6                           very much

Reliability                              very little                              1  2  3  4  5  6                          very much

Ambition                               very little                              1  2  3  4  5  6                          very much

Stability                                 very little                              1  2  3  4  5  6                          very much

Good character                     very little                              1  2  3  4  5  6                          very much

Likeability                             very little                              1  2  3  4  5  6                           very much

Wealth                                  very little                              1  2  3  4  5  6                           very much

Cleanliness                            very little                              1  2  3  4  5  6                           very much

Prestige                                 very little                              1  2  3  4  5  6                           very much

Adapted from Osgood, Suci, & Tannenbaum, (1957)