First stage: Identify the problem to be surveyed and set the target goals
This aspect involves identifying the main objective that drives the need for the survey process. The problem should be realistic and related to a major discipline such as academic, business, scientific discoveries just to mention but a few. A theory concerning the stated problem is formulated. It is drawn from the reference materials available for different scholars and researchers in the discipline that the survey is being conducted (Zhang 2000). In addition, the theory formulated should provide some of the facts that are known about the problem. In this stage, some of the previous surveys are reviewed by analyzing the recorded data in order to understand the requirements and improvements to make in the current survey. When the review process is complete, research goals and possible hypotheses are formulated. The research goals enable the person conducting the survey to focus onto the main objective of the survey. Hypotheses can either be null or positive and the choice of either depends with the with the surveyor’s preference. In a survey process, a hypothesis shows the surveyors thoughts of the possible solution to the problem being surveyed. For example, in a survey case about “Engineering students and alcohol addiction”, a possible hypothesis statement can be; “Most Engineering students are alcoholic.” In the same survey, a possible goal would be to identify the possible reasons that make Most Engineering students alcoholic.
Second stage: selecting survey method and the sampling technique
The second step identifies the target population from where to draw the data from. A sample is taken to represent the model of a larger group. Sampling can take the form of either probabilistic or non-probabilistic. Probability sampling also referred to as random sampling entails selecting a member of a general population has a non-zero chance of selection for the survey sample. However, non-probability sampling entails drawing a person in such a way that is practically impossible to associate a probability of selection in a random method to the any of the selected member in the population. Probability sampling method is highly recommended because it contains a reduced risk of sampling error and the confidence level gained is higher than for non-probability sampling technique. In a real survey case, Pew Research Center conducted a survey on the U.S public by using telephone numbers as the criteria for selecting the population sample in a method branded as “Random digit dialling” (Borkan 2004). Borkan (2004) argued that the method ensured all modile or landline numbers had the same likelihood of inclusion in the sample population. In addition, a small area may be isolated from a larger area and be used for the survey process. The results from the isolated area are taken as a representation of the entire area.
Third stage: draft a well-structured questionnaire
A questionnaire is a one of the three most applied instruments in a survey process. It enables the person conducting the survey process to exhaust all the possible areas of interest using set questions (Castle et al. 2005). This aspect is achieved by drafting questions that focus on the diverse information required about the area or subject of study. In this stage, it is important to consider the quality of the set questions because it directly determines the quality of the expected responses to the actual field survey (Terwee et al. 2007). Questions designed should accurately measure the sample population’s opinions, experiences, and their behaviours. In this quest, a person conducting the survey should consult previous surveys that have been conducted to determine the responses that are associated with a particular way of formulating questions (Larossi 2006).
Questionnaire development
First questionnaire drafting
The first draft of the questionnaire is developed from the topics that are associated with the survey. All questions are listed without considering the categories in which they will be put in the final draft. This aspect involves thinking about the happenings that are taking place in the nation where survey is being conducted and to an extent the entire world (Kozak, Owings and Hall 2005). Tracks of opinions that have evolved over a given period are reviewed to update them and make them fit for use in the current survey. Essentially, the first draft of the questionnaires is achieved through brainstorming of ideas by the group conducting the survey or by involving other specialists is not involved in any way in the current survey being undertaken (Jacobs et al. 1993). All questions raised during the drafting session are listed and discussed upon in order to check their general suitability in the survey. In this stage, only the questions that are not related to the topic of investigation are ruled out. For example in a survey process on the topic “cause of school drop-outs”, a question such as the one seeking to clarify information about the complexion of a student and how that is related dropping out would be ruled out in this stage. However, complex analysis of the proposed questions is done at the second screening of the questions.
Review questionnaire
This stage requires a team of survey experts to determine the worthiness of every question in relation to the survey topic. Every question earlier proposed is then pre-tested in order to check its fitness for the survey field work (Gosling et al. 2004). In this aspect, changes are made to the drafted questions that are either found to be ambiguous, repetitive or wrongly formatted. It is inappropriate to use a wide range of questions in the survey; therefore, the survey is limited to the scope contained in the topic of study. The guiding parameters in selecting the most appropriate questions to adopt are strictly guided by the scope and aim of the survey process.
Second drafting of questionnaire
A well-designed questionnaire is then formulated using the set of questions proposed in the first questionnaire draft. The questions are then put into either open end, closed end or ranking categories. Consequently, it is advisable to list questions in the order of closed-end then ranking and lastly the open end ones in order to enhance the type of responses provided by the target population. Open-ended questions should be less because they tend to consume more respondents time. The choice of questions, however, is determined by the set objective(s). For example, if the survey to be conducted concerns “The education of the kindergarten learners”, it will be appropriate to narrow the topic further before designing the survey. One way of doing this is to consider the gender aspect and the when question (Collins 2003). In relation to this example, a possible question would be to determine the school starting age for kindergarten learners in a given area.
Code book preparation
A code book provides the basic format in which the questions proposed in the draft questionnaire will appear in the final questionnaire to be used for collecting data in the field. The book provides sections for gender, age, a title that the respondent holds and many other parameters. Therefore, the coding of the final draft should include as much information as possible in order to make the analysis of the collected simpler as compared to non-coded formats. In addition, coding involves selecting the most appropriate language and tools to use in the questions. In this aspect, key elements such as colour of the fonts and paper to use in the questionnaire are considered. It is recommended that the questionnaire proposed should take care of the slow learners, visually impaired, physically handicapped and other similar hindrances which may segregate a section of the selected in sample. In an attempt to reduce some of these draw backs, a proper coding is included in the questionnaire. Use of coloured boxes in the ranking and closed-end types of questions helps to alleviate a problem of legibility because the choices provided are legible and are presented in an appealing way. In such a format, only a tick is required as an appropriate response to the questions formulated. Most of the questionnaires that have been found to be effective have been shown to contain an element of creativity in terms of the language used and the formatting of the questions.
Determining the method(s) of analysis
The choice of the topic of survey influences the method of data analysis that will probably be used for the analysis of the survey data. In most cases, survey data is analyzed using either the quantitative or qualitative or even both approaches. In qualitative approach, data is analyzed in non-quantifiable elements such as quality characteristics associated with human beings (Child, Faulkner and Tallman 2005). For example, in a survey conducted to determine the voting blocks in USA, a possible method would be qualitative. In this example, the person conducting the survey can associate different states in the USA with the inhabitants found in the particular region. Therefore, the skin color would be the deterministic factor to associate the voting blocks to the number of people voting for a particular candidate. In quantitative approach, quantifiable mathematical elements of analyzing data are used. These mathematical approaches include elements such as mean, variations, and standard deviations among others. This method of data analysis seeks to establish more concrete relationships between sets of data, providing guidelines on how to formulate questions in the questionnaire. The choice of using either of the two approaches or both entirely depends on the problem being surveyed and some of the expected responses.
Pretesting the proposed questionnaire
The questionnaire that has successfully passed all the previous stages is then tested using some selected individuals in the team conducting a survey. This activity uses a few respondents and is meant to prepare the whole team for the actual survey activity. The chosen members fill several questionnaires in a mock survey activity conducted by the survey leaders. In this exercise, only data collection activity takes place whereby the questionnaire materials are harmonized further in order to get ready for the actual survey.
Mock analysis of the pre-testing activity
The results collected from the mock survey earlier conducted can be analyzed using one of the methods earlier highlighted. Pre-analysis activity is important in showing the survey team where to put more emphasis when the actual survey will be conducted. The analysis team should check the data reliability and validity measures and consequently, advise the survey team on the necessary actions that need to be taken.
Revision of the codebook
Owing to the proceedings of the mock survey and analysis, the codebook is revised further as per the analysis team directions. In this case, only minor details regarding the appearance of the questions are harmonized.
Drafting of the final questionnaire
In the preparation of the survey activity, the proposed questionnaire is reviewed as per the analysis of the mock survey. In this case, the structuring of questions and the time that has been allocated for answering is enhanced paving way for standardization of the entire document. The document is then proofread to correct any grammar error and sealed for use in the survey activity.
Prepare a guideline sheet for the interviewers
In order to enhance flow of survey activities in the field, the survey team leader should prepare a procedural guideline sheet for the interviewers. This sheet should only be used as a guide in administering of questions and for providing the necessary structural support for the survey tool (Andrews, Nonnecke and Preece 2003). The guideline sheets are important tools for the management of a survey process because it provides the necessary information to the person conducting the survey; giving the most important areas to follow in realizing a successful survey process. It would also be necessary to include some checks and balance to the survey team such as provision of extra material for the respondents who have additional information apart from the one contained in the questionnaire. Managing the survey team would entail advocating for team work between all the people involved. This achievement can be realized by creating a good rapport and ensuring all members are treated equally. Another way of promoting effective survey process would be to award tokens to both the survey team and the target population on a random selection but creating the awareness first, prior to the survey activity.
References
Andrews, D, Nonnecke, B & Preece, J 2003, ‘Electronic survey methodology: A case study in reaching hard-to-involve Internet users’, International Journal of Human-Computer Interaction, vol. 16 no. 2, pp. 185-210, Web.
Borkan, J 2004, Mixed Method Study: A Foundation of Primary Care Research, Annals of Family Medicine, vol. 2 no. 1, pp. 4-6, Web.
Castle, N, Brown, J, Hepner, K, & Hays, R 2005, ‘Review of the literature on survey instruments used to collect data on hospital patients’ perceptions of care’, Health services research, vol. 40 no. 6, pp. 1996-2017, Web.
Child, J, Faulkner, D & Tallman, S 2005, ‘Cooperative strategy: Managing alliances, networks, and joint ventures’, OUP Catalogue, vol. 4 no. 16, pp. 19-27, Web.
Collins, D 2003, ‘Pretesting survey instruments: an overview of cognitive methods’, Quality of Life Research, vol. 12 no. 3, pp. 229-238, Web.
Gosling, S, Vazire, S, Srivastava, S, & John, O 2004, ‘Should we trust web-based studies? A comparative analysis of six preconceptions about internet questionnaires’, American Psychologist, vol. 59 no. 2, pp. 93, Web.
Jacobs D, Ainsworth, B, Hartman, T & Leon, A 1993, ‘A simultaneous evaluation of 10 commonly used physical activity questionnaires’, Medicine and science in sports and exercise, vol. 25 no. 1, pp. 81-91, Web.
Kozak, L, DeFrances, C & Hall, M 2006, ‘National hospital discharge survey: 2004 annual summary with detailed diagnosis and procedure data’, Vital and health statistics, vol. 162 no. 13, pp. 1-209, Web.
Kozak, L, Owings, M & Hall, M 2005, ‘National Hospital Discharge Survey: 2002 annual summary with detailed diagnosis and procedure data’, Vital and health statistics, vol. 158 no. 13, pp. 1-199, Web.
Larossi, G 2006, ‘The power of survey design: A user’s guide for managing surveys, interpreting results, and influencing respondents’, World Bank Publications, vol 13 no3, pp. 123-143, Web.
Terwee, C, Bot, S, DeBoer, M, VanderWindt, D, Knol, D, Dekker, J & DeVet, H 2007, ‘Quality criteria were proposed for measurement properties of health status questionnaires’ Journal of clinical epidemiology, vol. 60 no. 1, pp. 34-42, Web.
Zhang, Y 2000, ‘Using the Internet for survey research: A case study’, Journal of the American Society for Information Science, vol. 51 no. 1, pp. 57-68, Web.