Difference between Survey and Experiment

The following points are noteworthy so far as difference between survey and experiment is concerned:

• Surveys are conducted in case of descriptive research studies whereas experiments are a part of experimental research studies.
• Survey-type research studies usually have larger samples because the percentage of responses generally happens to be low, as low as 20 to 30%, especially in mailed questionnaire studies. Thus, the survey method gathers data from a relatively large number of cases at a time; it is essentially cross-sectional. As against this, experimental studies generally need small samples.
• Surveys are concerned with describing, recording, analysing and interpreting conditions that either exist or existed. The researcher does not manipulate the variable or arrange for events to happen. Surveys are only concerned with conditions or relationships that exist, opinions that are held, processes that are going on, effects that are evident or trends that are developing. They are primarily concerned with the present but at times do consider past events and influences as they relate to current conditions. Thus, in surveys, variables that exist or have already occurred are selected and observed. Experimental research provides a systematic and logical method for answering the question, “What will happen if this is done when certain variables are carefully controlled or manipulated?” In fact, deliberate manipulation is a part of the experimental method. In an experiment, the researcher measures the effects of an experiment which he conducts intentionally.
• Surveys are usually appropriate in case of social and behavioural sciences (because many types of behaviour that interest the researcher cannot be arranged in a realistic setting) whereas experiments are mostly an essential feature of physical and natural sciences.
• Surveys are an example of field research whereas experiments generally constitute an example of laboratory research.
• Surveys are concerned with hypothesis formulation and testing the analysis of the relationship between non-manipulated variables. Experimentation provides a method of hypothesis testing. After experimenters define a problem, they propose a hypothesis. They then test the hypothesis and confirm or disconfirm it in the light of the controlled variable relationship that they have observed. The confirmation or rejection is always stated in terms of probability rather than certainty. Experimentation, thus, is the most sophisticated, exacting and powerful method for discovering and developing an organised body of knowledge. The ultimate purpose of experimentation is to generalise the variable relationships so that they may be applied outside the laboratory to a wider population of interest.
• Surveys may either be census or sample surveys. They may also be classified as social surveys, economic surveys or public opinion surveys. Whatever be their type, the method of data collection happens to be either observation, or interview or questionnaire/opinionnaire or some projective technique(s). Case study method can as well be used. But in case of experiments, data are collected from several readings of experiments.
• In case of surveys, research design must be rigid, must make enough provision for protection against bias and must maximise reliability as the aim happens to be to obtain complete and accurate information. Research design in case of experimental studies, apart reducing bias and ensuring reliability, must permit drawing inferences about causality.
• Possible relationships between the data and the unknowns in the universe can be studied through surveys whereas experiments are meant to determine such relationships.
• Causal analysis is considered relatively more important in experiments where as in most social and business surveys our interest lies in understanding and controlling relationships between variables and as such correlation analysis is relatively more important in surveys.