Senegal Behavior Change Communication Research: Kaolack Endline Report

"The assumption is that if parents believe that home literacy and language activities can enhance the success of children in school, and if they have specific tools to use, they will develop more self-efficacy, that is to say, belief that they have a role to play in the children's success."

This endline report summarises the results from a 3-month pilot research activity in Kaolack, Senegal designed to test whether social and behaviour change communication (SBCC) strategies, frequently used in the health field, can bring about changes in family members' knowledge, attitudes, and behaviours related to their children learning to read. The goal was to examine how SBCC - radio broadcasts, community meetings, community theatre, and posters reinforcing a common set of messages - could be used to increase family involvement in directly supporting the amount of time children spend at home reading or practicing early literacy skills. As the evaluation found, at the end of the 3 months, family members in Kaolack demonstrated strong recall of the main messages of the campaign and were much more likely than control families to espouse beliefs supportive of their children learning to read.

As noted here, results from the Early Grade Reading Assessment (EGRA) and similar methodologies have shown "alarmingly poor levels of reading among students in the first few years of primary school - with many children not learning to read at all despite being enrolled in school for several years....[I]n places where classrooms may be overcrowded, school days short, and teacher and student absenteeism high, it is reasonable to ask whether in-school activities alone are sufficient to achieve the amount of improved learning needed."

The report explores SBCC, a basic tenet of which is that information is necessary, but not sufficient, to change behaviour. "A well-executed SBCC campaign begins with formative research to investigate the obstacles to and facilitators for change in the desired behavior, including the factors known to strongly influence the behavior, such as the attitudes, norms, and perceptions of self-efficacy of the target group. The Integrative Model, proposed by Fishbein (2000) and refined by Fishbein and Cappella (2006), brings together several commonly cited behavioral theories and serves to predict and explain behavior by illustrating relationships among the factors that influence whether or not a person performs a desirable behavior, such as hand washing; or stops an undesirable behavior, such as smoking (see Figure 1 [on page 4 of the report])." In addition to communication objectives, outlined in the report, designed to build parents' intention to perform the desired behaviours, the project also addressed the key environmental and skills/abilities barriers identified in the formative research. For the former, the project provided story books in Wolof and French that were available to children through mobile libraries managed at the community level. To address skills and abilities, a consultant was recruited to design and test literacy and language activities that parents could doat home, regardless of the parents' literacy level.

With funding from the United States Agency for International Development (USAID)'s Bureau for Economic Growth, Education, and the Environment (E3), the Education Data for Decision Making II (EdData II) project contractor, RTI International, undertook the proof-of-concept study. In collaboration with Associates in Research and Education for Development (ARED), a Senegalese non-governmental organisation managing a bilingual education initiative in 3 regions in Senegal, RTI developed the campaign: Liifantu (Reading). The campaign was implemented in one of those regions, Kaolack, with another region, Rufisque, serving as control.

The objectives of the campaign were to enhance families' perception of the value of reading, promote reading and literacy of children as a pleasure and a shared responsibility, and strengthen the confidence of parents in their ability to improve children's success in reading. The ultimate goal of the campaign would be to see children spending time every day reading and/or practicing their literacy skills, families providing an environment that supports those activities, and family members engaging in reading-related activities with their children.

To ensure an optimum mix of methods and media, channel selection was based on ARED's local knowledge of the existing information structures and habits in the community they sought to reach. Community radio was identified as the preferred form of mass media, since listenership is high. Both radio spots and a weekly radio programme were planned to take advantage of the synergy between the two. The radio programme also served to reinforce and promote the interpersonal communication that occurred during community meetings by featuring successful parents and children. Community theatre reinforced and magnified the messages by expanding on the characters and situations featured in the radio spots. Printed posters, banners, and flyers provided visual reinforcement and served to spur discussion within communities and create demand for parental interaction, because children saw the images and asked their parents about them. All of the interventions were unified with a single logo and slogan (the slogan translates from Wolof into English as: “Reading is the root of knowledge. Dear parents, read with me!"), so the effect of each component was amplified by the others. In addition, facilitators led community meetings during which they demonstrated and taught activities parents could do with their children at home to support reading, even if the parents were illiterate. The campaign also helped overcome the lack of reading materials in the community by providing stocks of books that children and families could borrow.

RTI partnered with Africa Consultants International (ACI) to measure the intervention's impact on family behaviour related to early literacy, using a household-based survey at baseline (April 2015) and endline (January 2016) in both Kaolack and Rufisque. At each of the 2 sites, 240 interviews were completed at baseline, and 240 were completed at endline, for a total of 960 interviews. The report outlines the analytical approach informing the research.

According to the aforementioned Integrative Model, which incorporates several SBCC theories of why people adopt or reject behaviors, people's beliefs influence their attitudes, norms, and self-efficacy, all of which influence a person's intention to change behaviour. For example, family members in Kaolack were almost 9 times as likely as those in the control region (Rufisque) to say that a parent could read with his/her child as one way to support that child learning to read - showing a dramatic change in a key belief that influences attitudes and thus predisposes them to form the intention to help their child. Family members in Kaolack were 1.6 times as likely as control families to say they felt they could help their child learn to read, a perceived behavioural control (PBC) belief that directly influences self-efficacy. Conversely, control household members were 5 times as likely to say that being illiterate was an obstacle to helping their child, showing a negative PBC belief. Family members in Kaolack reported engaging in the desired reading-supportive behaviours at a much higher rate than those in the control region. For example, when respondents were asked if they helped their early-grade child with school, those in Kaolack were 3.2 times as likely to say yes, compared to those in Rufisque. When asked if anyone in the household helped the child, those in Kaolack were 5.2 times as likely to say yes. Those who said they helped their child were asked how they help; Kaolack respondents were 2.9 times as likely to cite "I ask the child to read aloud" and 2.6 times as likely to say that someone in the household listens to the child read. Besides listening to the child read, the literacy activities promoted in the community were the main actions available to illiterate parents. Participation in "causeries", during which the activities were demonstrated, was high (79%), and the vast majority of those who attended said they learned activities, did them with children at home, received cards to help them remember the activities, and were able to show the cards to the interviewer.

In addition to the differences between treatment and control households, family members in the treatment region were more likely to demonstrate the targeted knowledge, attitudes, and behaviours if they reported greater exposure to the campaign activities. In other words, the more people heard the messages, the more likely they were to report the knowledge, attitudes, and behaviours supportive of their children learning to read.

"This study showed that the idea of using SBCC to promote family attitudes and behaviors that are supportive of early-grade reading is viable and should be explored further. To that end, RTI, under the EdData II project, is implementing the campaign in Rufisque (April-June 2016), using essentially the same components....A third round of surveys, slated for July, will measure the impact of the campaign in Rufisque and the sustainability of the effect in Kaolack approximately six months after the end of campaign activities." The report offers some suggestions for future research beyond this particular project, noting that "an important topic for further study will be whether increased parental involvement in early grade reading affects learning outcomes." They suggest that further research should be designed to test different combinations of campaign strategies - with attention to comparing the effectiveness and the costs of different combinations of campaign elements. Lastly,...[e]ffective campaigns have a relatively short shelf-life. Longer-term strategies will need to consider how and when to renew or change campaign messages, the sequencing of messages that best moves target audiences through the stages of behavior change, and what different audiences need to be targeted at different times (within the context of a comprehensive communications strategy)."