Survey Designs Used to collect descriptive data Study attitudes or values of a group Common practices of a group May be used to describe elements of a specific group or generalize to a larger population May study relationships between/among variables
Survey Methods Interviews Investigator asks specific questions of a respondent and records responses Affords opportunity for in-depth analysis Types Structured - investigator uses a specific set of questions Unstructured - no fixed agenda Has flexibility to diverge from the nature of a question vs. questionnaire “Go where the discussion leads you”
Survey Methods Questionnaires “Pen and paper response” a pre-determined set of questions Provides anonymity and encourages honest responses Questions should be standardized (everyone answers the same question) and determined to be valid and reliable Requires self-reporting & subject to bias recall
Questionnaires Disadvantage - may not be able to explore a response in depth This problem may be alleviated with a good design
Question Types Closed-ended Respondent selects from among several choices Example - simple structure (dichotomous response) “Are you currently practicing physical therapy?” a. Yes b. No
Question Types Response over a continuum “How important do you feel clinical research is to the practice of physical therapy?” a. Very important b. Important c. Somewhat important d. Not important e. Not sure
Question Types Branching 1. “Do you read a professional journal such as PHYSICAL THERAPY at least one time per month? A. No > Go to question 3. B. Yes 2. “How many hours do you spend reading this journal? _________ 3. ….
Question Types Rank-order “The following are characteristics of a researcher. Please order them in terms of importance from 1(most important) to 4 (least important.” _____methodical _____careful _____persistent _____on-time
Question Types Open-ended “Describe the characteristics that you feel are important to be an effective physical therapy practitioner.” _______________________________
Analysis Descriptive statistics Frequency analysis 65% of the respondents were female 35% of the respondents were male Frequency analysis Chi square examines significant relationships between variables