Health Education Research, Vol. 17, No. 2, 195-209,
April 2002
© 2002 Oxford University Press
Use of technology in reproductive health information designed for communities in South Africa
Medical Research Council, Private Bag X385,Pretoria 0001, South Africa
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The need to change sexual behavior to reduce the spread of the HIV/AIDS epidemic has highlighted the critical need for dissemination of effective reproductive health information. A wide range of media, including visual technologies, is used to provide health information to communities in South Africa. However, little is known about the ability of the technology used in communicating intended messages or on the comprehension of messages by the target audience. The study determined the ability of the technology to convey intended messages and the level of understanding of the information disseminated among the community members. A survey of both rural and urban communities of Umtata district in the Eastern Cape province was conducted to examine the readability and understanding of reproductive health information. The study revealed that urban communities understood disseminated reproductive health information better than rural. The findings also indicated that visual communication is much more problematic for older people. Frequency and consistency of reproductive health information improved the level of comprehension.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The rapid spread of the HIV/AIDS epidemic in South Africa has highlighted the critical need for individuals to change their sexual behaviors and therefore the need for effective dissemination of reproductive health information (RHI). The South African government and non-governmental organizations (NGOs) use a wide range of media, including visual technologies. Most households own radios and 56% of households own television sets, including some with more than one set (South Africa Television, 2000
). In the case of RHI, extensive use is made of written materials, including pamphlets, booklets, posters and flyers. These are generally developed and distributed mainly by the National Department of Health for the entire country. The materials are developed by in-house units or by private organizations through a tender process. In both cases the materials bear the name of the national department. South Africa has an extremely diverse population with different levels of education, different living conditions and a range of cultures. Although health information materials are generally pre-tested for aspects such as clarity, consistency, tone and appeal, concepts, language, comprehension, and relevance, this does not always ensure that they cater for the diversity of cultural concepts. Health information material is usually designed to appeal to a wide spectrum of communities. It is also often developed by urban-based health professionals who are sometimes insensitive to the diversity of the audience or the skills of consumers.
In South Africa, little attention has been given to the assessment of whether these efforts actually communicate the intended messages and whether they are understood by the target audience, but there are suggestions that messages are not communicated adequately especially on visual format (Hugo and Skibbe, 1991).
Qualitative evaluation of an AIDS health education poster revealed that messages on posters were not clear and needed to be explicit, direct and understandable (Evian et al., 1990).
A pilot study on visual literacy by Hugo and Skibbe showed variations in visual literacy skills among a group of illiterate women which indicated that extreme care must be taken in developing visual materials for health education (Hugo and Skibbe, 1991).
These studies cited in this paper which were conducted in South Africa provide evidence that there is a problem associated with visual materials used to provide information for individuals and communities in this country. This problem demands an intensive assessment on whether the visual technologies currently in use actually do convey the intended messages. There also appears to be a lack of a theoretical framework to guide the development of visual material for general audiences in South Africa. The assessment of the visual technologies in this paper is based on the author/message and audience/message relation model as explained in Sless (Sless, 1981) and Doak et al. (Doak et al., 1985).
There is a fundamental problem which seems to underlie the development of RHI in South Africa, i.e. vision and cognitive processes are separated from the cultural, environmental and socio-economic status of the individuals targeted by this type of information. This has also been the mistake committed very early in the history of visual communication when vision and understanding were understood as biologically distinct features (Kant, 1964; Blumer, 1967). The contemporary literature explains that visual communication and understanding (VCU) or decoding meaning from visual material is highly dependent on what is common between the visual material and the consumer of such material (Sless, 1981; Larkoff, 1984; Doak et al., 1985). Sless particularly argues that whenever the author of a message is not known by the audience intimately there is always an element of inference (Sless, 1981).
People will need to deduce the form and substance of the author from the message for them to understand the message. This inference occurs because the physical form of the message can exist independently of the author and the audience. This brings to light the importance of the relationship between the sender and the receiver in the process of decoding visual material.
There is a growing understanding that visual communication is popular but problematic. It is problematic because in understanding it the audience must regenerate the concepts used in developing it and in this process the message is redefined (Sless, 1981; Larkoff, 1984; Doak et al., 1985). It is important to realize that it is not the message itself that provides the meaning but the audience/author/message relationship. If there is no relationship between these aspects visual communication can be difficult to understand. There is a need to develop a close relationship between the developers of visual materials and the consumers of such material for them to be meaningful. A point to bear in mind is that visual communication takes place within socio-economic, cultural, developmental and environmental elements, which influence the response of different audiences. It also occurs within a popular culture of entertainment (television, films and drama) and decoration for some communities. For this reason people do not learn from them as argued by Sless that people have not been educated to learn from pictures (Sless, 1981). It can be even worse to expect them to learn and understand posters developed by health workers unilaterally who are culturally, structurally and professionally different from the target community.
In countries like South Africa where there are cultural differences which co-exist with inequitable socio-economic development understanding of RHI, posters can be problematic, especially if they are developed somewhere else.
Visual communication should be developed with better understanding of the cultural background, language and health belief models of the actual audience. For visual communication to be understood and effective it must be developed locally within the culture of consumers and the gap between developers and consumers must be narrowed to increase their ability to convey the intended messages.
The RHI materials assessed in this paper were developed by the National Department of Health and examined by a community that lives in the Umtata district, 1000 km from the place in which they were developed. As explained earlier, the National Department develops the materials and distributes them to health centers nationally. This paper examines these posters as well as other technologies used in RHI dissemination in an impoverished area of South Africa. It investigates their appropriateness and their ability to communicate information/ messages as well as the cognitive skills of the users of the reproductive health information. This study, based in one urban and one rural setting, attempts to assess the communication that is achieved by posters and pamphlet thereby comparing different technologies.
In this study, reproductive health information refers to information on HIV/AIDS, sexually transmitted diseases, sexuality, family planning and reproductive cancers.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Study population and sample
A population-based survey was done in an urban and a rural area of the Umtata magisterial district in the Eastern Cape province. This district was chosen because it falls in the most rural and under-served part of the country, and is 1000 km away from the center where the materials were constructed and pre-tested.
The area is generally characterized by poverty, high unemployment and poor infrastructure. Health services, including STD clinics, are mainly provided by the government and are generally overcrowded with inadequate medical and health equipment.
A table of random numbers was used to randomly select a village from the list of villages in the Umtata district. The selected village is typical of rural villages. Houses are thatched huts made of mud bricks. A residential area in the urban area was similarly selected from the list of areas from the Umtata municipal offices. The houses in this area are mainly small and made of commercial bricks with iron roofs.
A systematic sample of households in each area was identified by including every fifth household in a row. All the adult men and women aged 15 and above were included in order to minimize potential bias in the selection of eligible respondents by interviewers. The sample size was 831 adults.
Questionnaires
The questionnaire contained a section to assess the ability of RHI material to communicate intended messages in which VCU and frequency, consistency and understanding of reading materials were assessed. The second section contained socio-demographic details of the respondents. The questionnaire was also used to collect background variables and data on radio and television as these information provision channels were to be measured indirectly in the study.
The Wide Range Achievement (WRA), Cloze and Readability tests (Doak et al., 1985)
The WRA, Cloze and Readability tests were adopted and adapted for the study. The WRA test measures the ability to recognise words or posters, while the Cloze tests measures the ability to understand written messages. The Readability test measures the extent to which the material is readable to and comprehended by the readers (Doak et al., 1985). The WRA and Cloze tests were conducted by showing the poster to respondents and asking if they recognized and understood the meaning conveyed by the poster.
The Readability test was conducted for one pamphlet, which was available and written in the language (Xhosa) of the respondents. The Readability test was done by asking the respondent to read and summarize the meaning of the message in his/her own words. The Readability test was also used to measure the influence of frequency and consistency in understanding the RHI reading material (see Appendix for posters which were assessed).
|
After an introduction, the interviewers presented five posters on which all the text was covered. The posters were obtained from the health services in the area where the study was conducted, and included health promotion messages about HIV/AIDS and family planning. An adapted WRA and Cloze tests (Doak et al., 1985
) were applied by showing the posters to the respondents and asking them to explain what each poster was trying to convey. The interviewers then recorded a score for the average level of recognition and understanding of the five posters on the 0–10 scale where 0 meant that the respondent had said something far from the intended message and 10 meant the respondent had said the right message. After this, the same posters were presented with the written text visible.
Respondents were asked to examine the posters and indicate whether they thought there was relevance between the written messages and the picture on the posters. The interviewers gave respondents a five-line pamphlet about HIV/AIDS to read and were then asked to explain their understanding of the content. The pamphlet was in the vernacular and did not include any illustrations. It was the only pamphlet that was being used at the time of the study. The interviewers scored the respondents' understanding using the same 0–10 scale. The impact of frequency and the consistency of reading were assessed by asking respondents to read the pamphlet again and by scoring their understanding.
Socio-demographic details collected included age, gender, level of education, marital status, residence, previous exposure to reproductive health information on radio and television, talks at the health centers, and reading of reproductive health information.
Questionnaires were administered by 10 interviewers intensively trained by the researcher over a 3-day period. During the training the researcher ensured that the team of interviewers developed a common understanding of the intended messages in posters and pamphlet used in the study. Mock presentations and role-play were used to develop a standardized scoring system on the understanding obtained using the scale of 0–10. Mock interviews were conducted during the training to familiarize the interviewers with the questionnaire.
The study was conducted over 3 weeks. Interviews were conducted from morning to evening in both areas using Xhosa, the home language of both the respondents and interviewers.
A pilot of the instrument, including the scale, was conducted among community members from both the urban and rural communities before it was used. The information from the pilot was used to correct and modify the instruments.
Limitation of the methods
Scoring the understanding of the posters and pamphlet was problematic because it was determined by the interviewers' judgement. Scoring is subjective and difficult as a knowledge of the subject material is necessary before scores can be assigned. The field workers did not have an in-depth knowledge about reproductive health information but were only expected to be able to score reasonably accurately since the study needed a relatively superficial understanding of reproductive health to be able to measure understanding of the material used. For superficial levels of meaning and attitude, ratings, scores and rankings can be used with reasonable results (Oppenheim, 1992).
The scoring was based on the Readability, WRA and Cloze tests (Doak et al., 1985). Some of the respondents were excluded as the reading part of the instrument was not suitable for those who were illiterate. The validity and reliability of this adapted score is still to be evaluated. However, the overall reliability in the study has been ensured by intensive training and careful instruction in establishing standardized ratings.
While there were five posters that could be used, there was only one pamphlet available in the language of the respondent. This has limited the scope for assessing reading material.
Data analysis
The scores achieved in understanding of the posters and the first and second readings of the pamphlet are summarized by the median and interquartile range for place of residence by gender, age group, education category and marital status. Prior exposure to health centers and reproductive health information on radio and television are presented as percentages. A paired t-test was used to compare the poster score to the pamphlet score in the first and second readings.
People who had not attended a health center were unlikely to have seen such posters and were excluded from this comparison. A two-sample t-test was used to compare the mean score between the group who thought that there was or was not relevance between the caption and the image on the poster. Multiple regression analysis was used to assess the independent effects of the respective background variables on the poster score, the first reading score and the change in the score from first to second reading respectively. Significance testing was done at the 0.05 level of significance.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Background characteristics of the sample
The response rate was 100% (n = 831). Of the study population, 361 (43.4%) were from the urban area, while 470 (56.2%) were rural dwellers. The mean age was 28.4 years. There were more females (62%) than males (37.1%). There were more respondents within the age group 15–24 years (49.6%). There were 33.6% in the age group 25–40 years. Those above 40 years constituted only 16.2% of the sample.
The literacy rate (for this study refers to those with grade 7 and above) of the study population was (73.5%) which was an advantage for the issues under observation in the study. The educational average fell within grades 8–11. There were 37.2% respondents with grades 8–11 and 33.9% of the respondents had completed grade 12 and higher. There were 26.5% of the respondents who had completed grade 7 or less.
Prior exposure to reproductive health information
Table I presents results of those who had been exposed to information on reproductive health on the radio and television. A large proportion of the respondents reported that they had listened to reproductive health programmes on the radio (37.9%) in rural setting while only 6.6% reported that they had watched such programmes on television. Men watched such television programmes slightly more, as did those with a higher education level.
|
The rural community listens to the radio more than they watch television, which may be attributed to the fact that not many people in rural areas can afford television or have electricity. The majority (91.3%) of respondents reported that they had visited a Health Center previously. The proportion was slightly higher in the rural areas (93.4%) than in the urban area (88.4%).
Visual communication and understanding
Table II presents results of the understanding of the posters in quartiles, 25th percentile, median and 75th percentile, based on the scores recorded by interviewers. It can be seen that there were slight variations in understanding posters between urban and rural respondents.
|
Gender appears to influence poster understanding in the rural area but not in the urban setting. Rural men had a better understanding than rural women did. In rural areas, VCU is lower for those people above 40 years, while in urban areas it does not vary by age. Generally the findings reveal that education plays an important role in VCU. Those with higher education scored better than those with lower education, both in rural and urban settings. Those divorced, separated and widowed in the urban setting appear to understand visual communication more than groups with different marital status.
Further multivariate analysis on the poster was undertaken to assess the independent effects of these variables on the levels of understanding. The adjusted mean scores in Table III show that the higher the level of education, the better the understanding (P < 0.001) and that the respondents from the urban community had better understanding than those in the rural setting (P < 0.001).
|
Married respondents had better understanding than single respondents did (P = 0.02). Those respondents who reported that they had read RHI before had a better understanding (P = 0.015) while those who had listened to HIV/AIDS messages on television or at the Health Center did not. There was no difference in understanding between men and women or by age group.
Table IV shows that respondents detected a significant incongruence between the posters and their captions (P < 0.000).
|
Reading material and understanding (first reading)
The results in Table V
show median scores of both urban and rural respondents after the first reading of the pamphlet. Gender appears to have a slight influence on the understanding of HIV/AIDS reading material. In both the rural and the urban setting, women appear to understand reading material better than men do. The findings display variations in the understanding of reading material between the young and old. The older group appears to have lower levels of understanding than the younger groups and this is more marked in the rural setting (Figure 1). Understanding of the reading material appears to increase with levels of education in both the rural and the urban setting (Figure 2).
|
|
|
A multiple regression analysis for the first reading was done to identify the independent effects of the background variables. The results (Table VI
) reveal a marked education effect on the levels of understanding (P < 0.001) There is marginal significance for those previously exposed to reproductive health information with better levels of understanding achieved by those who had listened to reproductive health programmes on the radio (P = 0.045) and those who had read such information before (P = 0.046). Respondents who were separated showed lower levels of understanding than those who were single (P = 0.05).
|
Frequency of reading and understanding (first and second reading)
Table VII
presents results of the scores achieved by respondents after the second reading of the pamphlet. The second reading was done in order to assess the influence of frequency of exposure and consistency in the level of understanding the reading material. It appears that there is general improvement for most background variables, and especially for those with lower grades both in rural and urban settings.
|
Those over 40 years old in urban area had improved more compared to their counterparts in the rural setting. The difference in the scores between the first and the second reading is summarized in Table VIII
. It can be seen that there was an improvement between the first and second scores.
|
The level of improvement was further examined by using multiple regression analysis (Table IX
). The results reflect that the improvement is higher for men than women but this is not significant (P = 0.078). There is a vast difference in the improvement of comprehension in the urban setting when compared to the rural (P < 0.001). There is a much greater improvement in the young people aged 15–24 than in the older people aged over 40 years (P = 0.009). The least-educated group improved significantly more than the highest educated group (P = 0.004). Those respondents who were separated showed a greater improvement than those who were single did (P = 0.006) or those who were married (P = 0.022).
|
The study shows that previous exposure to a talk on RHI on television, on radio or at a health center does not affect the improvement made with a second reading of the pamphlet. Those respondents who had never read RHI before showed significantly greater improvement than those who had previously read RHI (P < 0.001).
Comparison between technologies for communicating RHI
The ability of the two technologies, posters versus pamphlets, in conveying intended messages was investigated and the results are shown in Table X based on a paired t-test. There was a significant gap between the mean scores for the poster and first reading of the pamphlet (P < 0.001) and between the poster and the second reading scores (P < 0.001).
|
The difference suggests that the posters are performing less well than the pamphlet in conveying intended information. The ability of television and radio to convey the intended information was indirectly assessed through their effects on understanding of the material used in the study. The results also reveal their poor performance as they did not have any influence in the understanding of the reading material on HIV/AIDS despite the fact that in Table I
the majority of respondents have listened to HIV/AIDS messages on the radio. |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The results of this study show that images on posters designed to inform people about reproductive health do not perform well in communicating the intended messages. Scores based on an adaptation of the Readability, WRA and the Cloze tests demonstrated low levels of understanding by respondents after being exposed to five posters which had their text concealed. On revealing the text, a high proportion of the respondents (81%) reported a lack of relevance between the image and the caption and text on the posters. It is interesting to note that those who did not perceive any relevance between the image and the text had achieved a higher level of understanding of the images than those who did perceive a relevance had. These results are commensurate with other studies which have shown some difficulty in correct identification of objects from pictures and the meaning of those objects (Sless, 1981
).
This study has investigated selected factors that may influence the level of understanding achieved through posters. Of those factors considered in the study, previous exposure to RHI was associated with better understanding. Those who had previous exposure have established a consumer information relationship which is an important step towards the understanding of information (Sless, 1981; Doak et al., 1985).
Higher education played an important role in understanding the messages on posters. This might be a result of more frequent exposure to RHI or the acquisition skills to obtain information from pictures through formal education. The results indicate a relationship between place of residence and understanding of visual communication. The urban dwellers had a higher level of understanding which might have resulted from seeing such posters in health facilities or from being exposed to more pictures and images in general. It is difficult to explain why single respondents should have less understanding than married or separated respondents but it is possible that they generally had less concern about RHI and therefore less background knowledge.
Sless suggests that it is important to distinguish the logical possibilities for using pictures and the everyday uses that people actually make of them (Slees, 1981). In the Umtata area, like many impoverished areas, people use pictures for decoration purposes or not at all. They do not usually use them for learning or for obtaining information.
The perceived irrelevance between the image and the caption might arise from the fact that people have basic expectations based on experience and environment. These are the basic expectations which should be examined when visual material is developed. For instance, developers of visual material should find out from consumers what is expected of a family planning poster before it is developed and take into account the differences between communities. Sless indicates that generalization about popular culture (visual communication) in any area has problems (Sless, 1981). These problems arise particularly if they are based on impression rather than carefully sifted evidence. There is clearly a need for more systematic evaluation of the role of posters in disseminating RHI and cultural influence in understanding them.
The basic questions that should be asked are: in which reproductive health message should posters be used?; for which target?; what common sources of visual communication are commonly available to rural communities?; etc. There is a strong need to develop posters within the culture of consumers to achieve better understanding.
Although the evaluation of reading material has been based on a single pamphlet, this study suggests that reading material communicates the intended information better than visual materials. While understanding of reading material was lower in the rural area and for the older respondents, the multivariate analysis showed that this was mostly explained by a strong relationship with education level. The important role played by the level of education in understanding of reading material is to be expected. The finding supports the view that understanding reading material depends on both the readability of the text and the educational level that a person has (de Saussure, 1974; Metz, 1977; Daok et al., 1985).
The study showed that understanding improved after a second reading, suggesting that frequency and consistency of messages is important in transferring messages. This was particularly the case for the urban dwellers, the young respondents, the less educated and those who had not previously been exposed to reproductive health information. The results are paradoxical because there was no association detected on improvement of scores in the second reading for those respondents who had watched television, or had listened to the radio or talks on reproductive health in health centers. A slight improvement was associated with those respondents who had no previous reading of HIV/AIDS information.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The study suggests that visual and audiovisual technology used to disseminate reproductive health information is not performing well in communicating messages in the study areas. Visual material, in particular, is failing to communicate intended messages, especially for rural communities that have low literacy skills. Further investigation is required to find out whether the posters failed at the level of transposition or transformation. If the problem is one of transposition, then the development of the material needs more attention than it is currently receiving. A problem of transformation reflects a lack of necessary skills to transform the poster on the side of users and would require that the users be empowered with the necessary skills to transform the information.
Frequency and consistency appears to improve understanding of RHI reading material. The findings suggest that communities need to have reading material at their disposal to read several times in order to understand information better. However, it is not known whether additional readings beyond a second reading would further improve understanding. Many respondents, particularly in the rural setting, did not achieve very high levels of understanding even after the second reading of the pamphlet. This reflects poor cognitive skills in the communities studied and highlights the importance of formal education to lay a foundation for the effective dissemination of health information.
Accordingly, caution must be exercised in drawing inferences beyond the immediate meaning of this data. The sample was drawn from a small, poorly developed area. However, as one of the very few studies to be conducted in South Africa on the efficacy of such tools, it has identified the failure of RHI materials to convey health messages to a rural and an urban community in an impoverished area.
Based on the 1996 census (The People of South Africa Population Census, 1996), nearly half the population of South Africa lives in the non-urban areas and an additional proportion live in impoverished conditions in urban areas. It is quite possible that initiatives to communicate health messages about reproductive health are failing in other areas of the country, which has very serious consequences in the light of the rapidly spreading HIV/AIDS epidemic.
|
Acknowledgments |
|---|
We acknowledge Dr Debbie Bradshaw for her comments on the earlier draft and Michelle Galloway for editing this paper, the community of Umtata district for their participation, and the interviewers who worked tirelessly to collect the data for the study.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Blumer, J. G. (1969) Producers' Attitude towards television coverage of an election campaign: a case study. In Halmos, P. (ed.), The sociology of Mass Media Communicators (The Sociological Review Monograph 1). University of Keele, Keele.
Coward, R. and John, E. (1977) Language and Materialism. Routledge & Kegan Paul, London.
de Saussure, F. (1974) Course in General Linguistics. Wade Baskin, Fontana/Collins, Glasgow.
Doak, C. C., Doak, L. G. and Root, J. H. (1985) Teaching Patients with Low literacy Skills. Lippincott, New York.
Evian, C. R., Ijsselmuiden, C. B., Padayachee, G. N. and Hurwitz, H. S. (1990) Qualitative evaluation of an AIDS health education poster: a rapid assessment method for health education materials. South African Medical Journal, 78, 517–520.
Hugo, J. and Skibbe, A. (1991) Facing visual illiteracy in South African Health education: A pilot study. Journal of Audiovisual Media in Medicine, 14(2), 47–50.
Larkoff, G. (1984) Classifiers as Reflection of Mind: A Cognitive Model Approach to Prototype Theory, Science Report 19. University of California Institute of Human Learning, Berkeley, CA.
Kant, I. (1964) Critique of Pure Reason. Norman Kemp Smith Macmillan, New York.
Metz, C. (1974) Film Language: A Semiotics of the Cinema. Oxford University Press, New York.
National Cancer Institute (1982) Developing Effective Print Materials for low Literate Readers. NCI, Bethesda, MD.
Oppenheim, A. N. (1992) Questionnaire Design, Interviewing and Attitude Measurement. Sage, London.
Sless, D. (1981) Learning and Visual Communication. Croom Helm, London.
South Africa Television (2000) Broadcasting Report. SAT, Johannesburg.
The People of South Africa Population Census (1996) Census in Brief Report no. 1:03-01-11. Statistics South Africa, Pretoria.
Received on August 15, 2000; accepted on May 16, 2001
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||