- KAP stands for knowledge, attitudes and practices: three distinct modules of the same questionnaire, measuring what people know, believe and do about a topic.
- The gap between the three, knowing but not believing, believing but not doing, is the diagnosis that decides the type of intervention: information, persuasion or removing practical barriers.
- A KAP study is usually conducted face to face, on a probability sample of the target community, with qualitative modules to understand the barriers.
- The instrument must be culturally and linguistically adapted, and knowledge questions have correct answers defined beforehand with specialists in the field.
- A baseline KAP, repeated at the end of the programme with the same instrument, is the basis of any evaluation of behaviour change.
1. What a KAP study is
A KAP study, from knowledge, attitudes, practices, is a quantitative survey on a representative sample of a population that measures three distinct things about a topic: what knowledge people have, what attitudes they hold towards it, and what practices, that is, concrete behaviours, they actually have. The topic may be vaccination, child nutrition, domestic violence, waste management, consumer rights, electoral participation, road safety, digital public services or anything else about which an organisation intends to change something in a community's knowledge, beliefs or behaviour.
The method was developed in public health and is today the standard instrument of international organisations, government agencies and NGOs in the programme design phase. It is a type of social survey with a fixed structure, and its strength lies in that structure: the three modules are measured separately and compared, and the comparison is the diagnosis.
2. Why three modules and not one
Social programmes frequently fail because of a single design error: they assume the problem is a lack of information. An information campaign is run, people learn, and nothing changes. The KAP study prevents this error, because it shows, before the programme, where the break actually is.
| Situation in the data | Diagnosis | Type of intervention |
|---|---|---|
| Low knowledge, unformed attitudes, absent practices | The topic is unknown | Information: campaign, materials, education |
| Good knowledge, negative attitudes, absent practices | People know, but do not believe or accept | Persuasion: credible messengers, social norms, addressing the fears identified |
| Good knowledge, positive attitudes, absent practices | People know and want to, but cannot | Removing barriers: access, cost, distance, time, services |
| Low knowledge, practices present | The behaviour exists for other reasons, not from understanding | Reinforcement and correction of wrong practices |
| All good on average, but one group lags behind | The problem is concentrated | Targeted intervention on the group identified |
Each row of the table calls for a different programme, with a different budget and different partners. Without KAP, the organisation chooses the row it assumes; with KAP, it chooses the row it measures.
3. Building the three modules
Knowledge is measured through questions with a correct answer: what causes the disease, from what age vaccination is given, where a complaint is filed, what a symbol on a label means. The correct answers are established beforehand with specialists in the field, and the questions are worded so as not to suggest the answer. A knowledge score is built from the answers, with thresholds of "adequate knowledge" defined and justified in the report.
Attitudes are measured through statements the respondent agrees or disagrees with, on 4-5 point scales: "vaccines are safe", "violence in the family is a private matter", "it is worth separating waste even if the neighbours don't". The statements include beliefs, risk perceptions, perceived social norms, that is, what people think others do and believe, and trust in the relevant institutions and sources. Neutral wording and a balance between positive and negative statements are essential.
Practices are measured through questions about concrete, recent and verifiable behaviour: "when was the last time you…", "in the last 30 days did you…", "where did you go when…". People are not asked "do you usually…", which produces the desirable answer, but about specific episodes. For sensitive behaviours, techniques for reducing social desirability are used: indirect questions, self-completion on the tablet, wording that normalises the behaviour. The module also includes stated barriers: why not, what would have to change.
4. Population, sample and method
The population of a KAP study is the community the programme targets: the residents of certain districts, women of reproductive age, parents of children under five, young people aged 15-24, households in a rural area. The sample is a probability sample, stratified by the criteria relevant to the programme, usually region, urban or rural residence, gender and age, and sized by the subgroups the programme will treat differently. A typical KAP has 600-1,200 respondents; when the programme compares districts or vulnerable groups, each needs at least 150-200.
The standard method is the face-to-face interview, because the populations targeted by social programmes are precisely those with weaker telephone and online coverage, because the questionnaire takes 30-45 minutes, and because the topics are often sensitive. Interviews are conducted in the respondent's language, Romanian or Russian, with a translated and back-translated instrument, and for specific communities, with interviewers from the community, specially trained. The multilingual team service covers this requirement.
5. The qualitative component
The survey measures how much; it does not explain why. That is why a complete KAP study includes a qualitative component, usually before the survey, to build the instrument correctly, and sometimes after, to explain the results: 4-8 focus groups with community members, separated by the relevant categories, and 10-20 interviews with key informants: health or social workers, teachers, local leaders, authorities, service providers. The qualitative component identifies the real barriers, the community's language, the messengers it trusts and the myths in circulation, that is, exactly what the questionnaire must measure and what the programme must address.
6. Cultural adaptation and instrument validation
A KAP questionnaire taken from an international guide is not translated; it is adapted. The knowledge questions must match the recommendations and services that exist in the country, not those elsewhere. The attitude statements must be worded in the terms in which the community thinks about the topic, identified in the qualitative phase. The practices must be those possible in the local context. The instrument is translated, back-translated, discussed with local specialists and piloted on 20-30 respondents from the community, and the pilot checks not only duration and comprehension but also whether the questions are culturally acceptable and whether they produce variation: a question everyone answers the same way measures nothing.
7. Analysis and report
The report of a KAP study presents: the knowledge, attitude and practice scores, in total and by subgroup, with the significant differences; the analysis of the link between the three, that is, where each step is lost; the barriers stated and those identified qualitatively; the information sources used and those trusted, which become the programme's channels; and the profile of the groups with low scores, which become the priority target. The recommendations section translates the diagnosis into programme design: which type of intervention, for whom, through which channels, with which messages and with which monitoring indicators.
All indicators are defined so that they can be measured again, identically, at the end of the programme. A baseline KAP followed by an endline KAP, with the same instrument and the same method, is the standard design for evaluating change, described in the guide Baseline, endline and impact evaluation.
8. Frequent mistakes
- The instrument translated, not adapted. Questions about services that do not exist locally or in terms the community does not use.
- Practices measured through stated habits, not concrete episodes; the result is what people think they should do.
- Correct answers not defined before fieldwork, set afterwards according to what came out.
- A sample of beneficiaries already reached by the programme, or of those easy to find at the community centre; the result describes those who do not need the programme.
- No qualitative component; the barriers remain assumed.
- Indicators that cannot be repeated; the final evaluation has nothing to compare against.
9. What to ask the agency
- The definition of the target population and the probability sample, with the programme's subgroups and the size of each.
- The prior qualitative phase and how it feeds the instrument.
- The list of correct answers to the knowledge questions, validated with specialists, before fieldwork.
- Measurement of practices through concrete episodes and the techniques for sensitive behaviours.
- Cultural adaptation, translation, back-translation and the pilot in the community.
- The face-to-face method, the interviewers and their training on the specific topic, the ethics and consent protocol.
- Analysis of the knowledge-attitude-practice link per person, not only separate scores.
- Indicators defined for repetition at endline and recommendations formulated as programme design.
The KAP study is the instrument through which a social programme stops being an assumption about a community and becomes a response to what the community knows, believes and does. It costs a fraction of the programme and decides, beforehand, whether the money goes into information, persuasion or removing barriers. For data collection, see the CAPI and CATI page.
See also: Social and political surveys CAPI and CATI Face-to-face interviews (CAPI) Baseline, endline and impact evaluation How focus groups work Multilingual team
Do you have a decision to take?
Tell us the problem, not the method. We propose the right research design and explain what it can and cannot deliver.