Innovations in experimental design in attitude surveys
The word survey is used most often to describe a method of data collection from a sample of individuals. A survey can be said to be an activity that is aimed at collecting information in an organized and methodical way about characteristics a phenomena from some or all groups of a population using concepts that are well defined methods, concepts and procedures, and compiles the summary into useful form.A survey consists of several interconnected steps that include: defining objectives, choosing a sample frame, selecting the sample design, designing questionnaires, collecting data, processing and analyzing data, and disseminating the data and documenting the results of the survey. It takes place in phases, starting with the planning phase, followed by the design and development phase and finally the implementation phase. The entire survey process is then reviewed and evaluated.Survey experiment continues to play a pivotal role in public opinion research. By using this methodology, we can gain more advantage we can study public opinion and political psychology.
The survey has to be carried out step by step, following specific procedures and formulas if one wants to yield meaningful and accurate information. To understand the entire process, it is important to understand individual procedure and how they are related and interconnected.
Uses of Survey
The survey is used to collect information where no data or sufficient data exist. Sometimes this need arises from within statistical agencies or when a client wants to study certain characteristics of the population, build a database for analytical purposes or to test a hypothesis. Most information we have comes from cross-sectional data that has been statistically analyzed and to a lower extent, panel survey data. For a long time, scholars have relied on this data to explain a variety of phenomena that include: policy preference, economic assessments, voting decisions and, candidate evaluations, among others (Brady and David, 2000).
Political scientists rely on survey experiments to identify how the public make decisions and respond to political objects, so as to better understand politics. When survey is used with representative samples, survey experiments can provide firmly grounded inferences about people’s attitudes and behavior. In general, survey experiments are considered to be the best type of research design when making causal references, avoiding issues of reverse causation, measurement error and omitted variable bias (Holland, 1986).
Types of Survey
There are various types of survey research depending on the instruments used to collect data and the duration over which data is collected before the survey is completed. The various types of survey research instruments are questionnaires and interviews. Questionnaires are utilized in various survey methods and may be self-administered, or group administered. The major types of survey are:
This type of survey involves thecollection of data at a single period. Cross-sectional surveys normally use questionnaires to ask about a given topic at one point in time. For example, a researcher can conduct a cross-sectional survey asking young boys on their views on cigarette smoking as of June 2010.
In this type of survey, the researcher attempts to collect information over a given period or from a given period to another. Longitudinal survey aims to collect data and examine changes in the information collected. Longitudinal surveys are employed in panel studies, cohort studies and trend studies.
Is birth order a factor in determining parental attention and academic achievement?
30 participants will be analyzed and will be categorized into firstborns, middleboxes and lastborns. Each participant must have completed a school academic year 2014-2015 and will have their grade average established. To be included in the study, each participant is required to have living parents and at least a sibling. The survey will be distributed to a sample of high school and college aged individuals in families consisting of two parents and two or two or more children. The study will target participants aged between 15 and 25. This is to increase the chances that they will be residing with their parents. Data will be then analyzed using correlation between the variables thatare total attention score, birth order, and academic achievement score. Individual results will be divided into three groups as per birth order (firstborn, middleborn and last born. The academic achievement and attention score will be compared using the mean of each birth group.
The participants in the survey will be categorized into three sections: the first child will be considered as firstborn; the last child was the lastborn, and all the children in between will be considered middleborn. Each participant will be given an anonymous survey enclosed in an envelope, to ensure confidentiality and get honest responses. Surveys will be conducted in person to ensure any questions participants may have are addressed.
The variables that will be manipulatedaretotal attention score, birth order, and academic achievement score. The independent variable is the birth order while the dependent variable is the academic achievement score and attention score.By manipulating these variables, we can get appropriate results to get answers to the research question (Sniderman and Grob, 1996).
Brady, Henry E., and David Collier, eds. (2000).Rethinking social inquiry: Diverse tools, shared standards. Lanham, MD: Rowman and Littlefield.
Sniderman, P. M., &Grob, D. B. (1996).Innovations in experimental design in attitude surveys.Annual Review of Sociology, 22, 377–399.
Holland, P. W. (1986). Statistics and causal inference.Journal of the American Statistical Association, 81(396), 945–960.