Attitude and opinion research requires considerable planning, careful design of survey procedures and sampling techniques, and practical ways of collecting, analyzing, and reporting results. The following suggestions are based on the extensive public opinion research conducted by the University of Alabama. (1) Decide what specific problem is to be solved and whether a survey is the best way to solve it. (2) State the survey objectives clearly in writing. (3) Define the survey population. (4) Allow enough time to plan, conduct, and report the survey. (5) Identify and eliminate biases. (6) Be aware of politics involved. (7) Do not hesitate to ask others for assistance. (8) Budget costs to assure an accurate survey. (9) Recognize survey limitations. (10) Provide for "no opinion" responses. (11) Determine advantages and disadvantages among mailed questionnaires, personal interviews, and telephone surveys. (12) Develop the survey instrument in terms of the type of survey and the budget. (13) Determine the survey procedure in terms of the entire population or a random sample. (14) Design a brief, specific, concise questionnaire using input from those to be surveyed. (15) Pre-test the questionnaire with groups similar to those to be surveyed. (16) Provide a stamped, self-addressed envelope with mailed questionnaires. (17) Tabulate the results as soon as possible and release them. A selected bibliography is appended. (MB)