Background
Historically, data collection relied on paper-based methods like surveys, questionnaires, physical observations and other traditional methods. This approach is time-consuming, prone to errors while collecting data and during data entry, limited in the type of data that could be collected as large-scale surveys may be very tasking. The widespread use of computers and the internet led to the development of digital data collection methods.
Computer Assisted Interviewing (CAI)
Computer Assisted Interviewing (CAI), otherwise known as electronic or digital data collection, is the process of collecting data electronically using existing technology such as laptops, smartphones, tablets, and other digital devices. The rise of mobile devices with sensors further expanded digital data collection possibilities, allowing for real-time data capture on location, activity levels, and even health information. In contrast to paper-based methods, digital data collection methods have some generic advantages which include: enhanced accuracy and efficiency, increased accessibility and reach, improved data quality and organization, real-time insights and analytics, cost-effectiveness, among others.
CAI is one of the major breakthroughs in survey data collections. It was first implemented in 1990 on a Labour Force Survey in the United Kingdoms.
Figure 1: R-DATS Digital Survey Process
Some of the commonly used electronic data collection softwares at R-DATS include ODK, Survey CTO, Survey Solutions, Kobo Collect, CS-Pro and CommCare, among others.
Overview of general advantages and disadvantages of CAI
Advantages | Disadvantages |
If well programmed, it reduces both the possibility of data entry and enumerator’ errors. It can save survey cost such that survey results can be generated quickly and there is no need for additional data entry after data collection in the field. | Training of enumerators would include specific module on the use of CAI devices and software. CAI also requires support of experienced digital survey scripters with some fundamentals in computer programming. I.T. related issues can be experienced. |
The questions are progressive, so interviewers cannot jump questions or ask the wrong questions. It contributes to environmental sustainability by reducing paper usage. | If CAI data gets corrupted before it is uploaded to a server, it is lost forever. |
Questionnaires can be improved in between interviews. It allows for electronic high frequency checks (HFC) during data collection and identified areas for improvement (including errors and adjustment) can be implemented in the instruments while in the field. Mathematical calculations can be carried out within the digital program. | Reluctant to participate in the survey due to suspicion of CAI devices (such as laptop or tablet etc.) by certain categories of respondents. This may affect the survey responses and coverage, if not well managed. |
The data is ready for analysis. Analyze as soon as data collection is completed. | Researchers may encounter difficulty in relating with printed CAI-programmed questionnaire. |
The interviewing process is made quicker because data is entered as it is obtained. Since feedback can be collected live, researchers and analysts can interrogate the survey dataset while fieldwork is still taking place. This also includes reporting survey results within live dashboards. | Data collection and data entry are two separate activities with the latter taking place after the former. |
It can automatically record each interview’s start time, end time and GPS location, making it easy for supervisors or data manager to check whether an enumerator indeed conducted a given interview or not. |
Modes Of Digital Data Collection
In today’s information-driven world, gathering accurate and insightful data is crucial for informed decision-making. But with the shift from paper to electronic devices, how exactly do we collect data digitally? This blog post unveils the various modes of digital data collection, namely: CAPI, CATI, CAWI and CASI, each with its unique strengths and weaknesses. Details about each of these are provided below.
- CAPI: Computer-Assisted Personal Interviews (CAPI) is a face-to-face (in-person) data collection method in which the interviewer uses a tablet, mobile phone or a computer to record responses given during the interview directly into a computer system.
Strengths and Weaknesses of CAPI
Strengths | Weaknesses |
Offline data collection is possible. Improved data storage and security system is available | Enumerators require extra training on use of CAPI device and software. |
Real time monitoring of survey progress during data collection is possible. | Limited to in-person surveys. |
If well programmed, it reduces data collection and data entry errors. | CAPI surveys may be difficult to conduct in crime-ridden areas, since tablets or phones are easy targets for thieves. |
It can automatically record each interview’s start time, end time and GPS location, making it easy for supervisors to check whether an enumerator indeed conducted a given interview or not. |
2. CATI: CATI stands for Computer Assisted Telephone Interviewing. It is a research method that conducts survey research in a telephonic manner guided by a computer. The interviewer makes a call to the respondent and conducts the whole research over the telephone by using a script that is displayed on a computer. Computer-Assisted Telephone Interviewing (CATI) is a valuable tool for collecting high-quality, reliable data efficiently. As technology continues to evolve, CATI systems are becoming more sophisticated, offering features like multilingual support and integration with other data collection methods.
Strengths and Weaknesses of CATI
Strengths | Weaknesses |
Real time monitoring of survey progress during data collection is possible. Offline data collection is possible. | Enumerators require extra training on use of CATI device and software, including case management system (where applicable). |
Improved data storage and security system is available | Difficulty in reaching respondents located in areas with no or poor telecommunication services. Respondents with no telephone contact (direct or indirect) may be ineligible to participate. |
Interviews can occur at a more convenient location and time for the respondents. | With telephonic interviews for surveys, you cannot expect 100% loyal answers every time. |
It can automatically record each interview’s start time, end time and GPS location, making it easy for supervisors to check whether an enumerator indeed conducted a given interview or not. | Also, several call backs and interview re-scheduling may be expected. It may incur a higher cost, depending on the contexts. |
To ensure good responses, it is not suitable for longer survey length. Prior studies in development research suggest that phone surveys that are longer than 20 minutes may result in lower response rates. | |
It may result in high incomplete interviews or lower survey coverage, depending on the nature of the study. | |
Coping with open-ended responses presents some problems on CATI: Although the systems can accommodate open-ended responses, capturing them requires interviewers to have fast and accurate typing skills. |
3. CAWI: Computer-assisted web interviewing (CAWI) is an Internet surveying technique in which the respondent follows a script provided in a website.In CAWI, respondents access surveys or questionnaires via a web-based platform, the respondents then fills out the survey at their convenience, using their own device (smartphone, tablet, or computer). Some CAWI systems also provide real-time monitoring, allowing researchers to track the progress and completion of surveys. With Computer-assisted web interviewing, respondents can participate in the surveys from anywhere in the world.
Strengths and Weaknesses of CAWI
Strengths | Weaknesses |
Relatively lower data collection costs because it is self-administered online by the respondents. | More suitable for survey targeting high literate respondents with access to web browsers, mostly with their own gadgets. With CAWI surveys, you cannot expect 100% loyal answers every time. |
Suitable for swiftly and comfortably reaching larger respondents at a time, regardless of their locations around the world. | To ensure good responses and completion rate, it is not suitable for longer survey length |
Since it is self-administered, there is no need to hire enumerators. | Access to internet is essential. CAWI surveys are prone to technical issues, such as site crashes or internet connectivity issues. This can increase the survey abandonment rate, reducing the sample size and potentially introducing an element of bias |
4. CASI: CASI stands for Computer-Assisted Self Interviewing. In a CASI survey, the respondents use a computer or tablet to complete a survey on their own. The respondent reads the questions on the screen and enters their answers directly into the computer. This approach enhances the efficiency, accuracy, and confidentiality of data collection. The success of CASI depends on the target audience’s digital literacy and access to technology. It also requires robust infrastructure to prevent issues like data loss or incomplete responses during interruptions
Strengths and Weaknesses of CASI
Strengths | Weaknesses |
It is self-administered and/or self-entered survey data. Hence, there is no need to hire enumerators. | More suitable for survey targeting high literate respondents. Likewise, the survey is likely to attract only respondents who are computer literate with good access to internet connectivity. Absence of all these may lead to biased samples or underrepresentation of certain demographics (i.e. less educated and older people, etc.). |
It ensures higher anonymity, thereby increases chances of collecting more honest responses especially on sensitive topics | The survey can miss feedback and clarification/quality control that a personal interviewer could provide especially when completed by less literate respondents. |
KEY FEATURES OF DIGITAL DATA COLLECTION TECHNIQUES
Features | CAPI | CATI | CAWI | CASI |
Administration | Interviewer-administered | Interviewer-administered | Self-administered | Interviewer/self-administered |
Quality | Increased accuracy, improved data quality. | Increased accuracy, improved data quality | Increased accuracy, improved data quality. | Increased accuracy, improved data quality. |
Costs | Relatively high cost | Relatively high cost | Reduced cost | Relatively high cost |
Privacy | Relatively reduced privacy | Relatively reduced privacy | Increased privacy | Increased privacy |
Data security | Highly secured | Highly secured | Highly secured | Highly secured |
Conclusion/Summary
Digital data collection methods have revolutionized the process of gathering, analyzing, and leveraging data, making it more efficient, accurate, and adaptable to various research needs. Each of the techniques, i.e. CAPI, CATI, CAWI, and CASI, offer unique advantages and cater to diverse research contexts, from face-to-face interviews to web-based surveys.
Digital or electronic data collection techniques offer improved efficiency and effectiveness in data collection management by eliminating use of paper-based questionnaires, enhancing data quality through timely validation checks, providing real-time accessibility for immediate analysis and feedback, and data security through encryption and compliance with privacy regulations. It also allows collection of data in multimedia formats such as videos, images, audios etc., and contributes to environmental sustainability by reducing paper usage.
Despite its identified shortcomings, electronic data collection methods are widely accepted as modern best practices while efforts are being made to improve on its usability, adaptivity and operationalizations in different research contexts and locations, machine learning, and advanced data analytics, continue to improve these methods. These innovations make digital tools more adaptable and resilient, enabling them to meet the diverse needs of researchers across different sectors. As the demand for reliable data grows, digital data collection will continue to play a pivotal role in shaping informed decisions and driving impactful solutions.
R-DATS Consulting uses advanced and innovative research, evaluation and statistical methods to gather high-quality data and generate strong evidence for impact and improved policy decisions. We achieve this using innovative and advanced research, evaluation, data analysis, data management and collection methods including CAPI, CATI, CAWI and CASI.
Resources and further reading:
- (2024, April 30). Introduction to the role of OECD DAC criteria in impact evaluation. ActivityInfo. https://www.activityinfo.org/blog/posts/2024-04-30-Introduction-to-the-role-of-oecd-dac-criteria-in-impact-evaluation.html
- The World Bank. Computer-Assisted Personal Interviews (CAPI). https://dimewiki.worldbank.org/Computer-Assisted_Personal_Interviews_(CAPI). Date accessed:17 September 2024.
- ProGlobal Business Solutions. CAPI, CATI, and CAWI Research Methodologies. ProGlobal Business Solutions. https://www.proglobalbusinesssolutions.com/capi-cati-cawi-research-methodologies/. Date accessed: 22 November 2024.
- B2B Internationa. What is CATI? B2B International https://www.b2binternational.com/research/methods/faq/what-is-cati/. Date accessed: 14th September 2024.
- DJS Research. Computer-Aided Self-Interview (CASI). DJS Research. https://www.djsresearch.co.uk/glossary/item/Computer-Aided-Self-Interview-CASI. 14th September 2024.
- MBA Skool. Computer-Assisted interviewing. MBA Skool. https://www.mbaskool.com/business-concepts/marketing-and-strategy-terms/17845-computer-assisted-interviewing.html. 14th September 2024
