Study Guides (248,357)
Canada (121,502)
Psychology (1,882)
PSYB01H3 (82)
Anna Nagy (35)
Final

psyb01 final exam.docx

13 Pages
196 Views
Unlock Document

Department
Psychology
Course
PSYB01H3
Professor
Anna Nagy
Semester
Fall

Description
Observational Methods Chapter 6 QUANTITATIVE AND QUALITATIVE APPROACHES  Empirical/ Quantitative approach (most common) developing theories, generating hypotheses to test those hypotheses, including operationally defining variables; collecting numerical data from many participants and analyzing those data using statistics  Qualitative Approach deep description of ppl’s behavior in natural settings - an example question would be: how the lives of teenagers are affected by working outside of school - you would then develop a questionnaire and assign numerical values to the responses, then analyze the data using statistics - the description of the results may include the percentage of teenagers who work and the way this percentage varies by academic average NATURALSTIC OBSERVATION Description and Interpretation of Data Issues in Naturalistic Observation Participation and Concealment Defining the Scope of the Observation Limits of Naturalistic Observation SYSTEMATIC OBSERVATION Coding systems Issues in Systematic Observation Reliability Reactivity Sampling CASE STUDIES ARCHIVAL RESEARCH Statistical Records Survey Archives Written and Mass Communication Records Content Analysis: Making Coding Systems for Documents Surveys Chapter 7 ASKING PEOPLE ABOUT THEMSELVES: SURVEY RESEARCH - survey research: uses questionnaires and interviews to ask ppl to provide info about themselves (attitudes, beliefs, demographics (age, sex, income, marital status), other facts, past/intended future behaviors - this method of collecting data is used in both qualitative and quantitative research  Why Conduct Surveys - important way to study relationships among variables and ways that attitudes and behaviors change over time or among diff groups of ppl - survey research is also important as a complement to experimental research findings (starting point)  multiple methods are needed to understand any behavior - we assume that ppl will provide accurate/truthful info when filling out a survey - this issue is addressed by studying possible biases in the way ppl respond - a response set is a tendency to respond to all questions from a particular perspective rather than to provide answers that are directly related to the questions, reducing the usefulness of data  most common is called the social desirability/”faking good” which leads the person to answer in the most socially acceptable way (reflects most favorably on the person)  most acute when the topic is violence, aggressive behavior, substance abuse, sexual practices, some scales measure how much a person is trying to present themselves as favorable  Paulhus, Harms, Bruce and Lysy (2003) operationally defined social desirable response bias as the extent to which ppl claim familiarity with fake “facts” – e.g. “plates of parallax” is a real term” – that are hidden in longer lists of true facts – eg. “that nuclear fusion” is a real term” – across a variety of domains (physics domains uses in this e.g.)  ppl most likely misrepresent themselves when you are not fully informed about the purpose/use of the experiment, and is not given feedback about results, assures confidentiality  Constructing Questions to Ask a. Defining the Research Objectives First step: explicitly determine research objective: what do you want to know Second step: choose survey questions that are tied to the research questions that are being addressed These are the 3 general types of survey questions: I. attitudes and beliefs - focus on the ways that ppl evaluate and think about issues - e.g. should more money be spent on mental health issues - e.g. did police respond to your call appropriately II. facts and demographics (age, gender) - indicate things they know about themselves and their situation - e.g. consumer reports on repairs to products such as cars, dishwashers - e.g. ethnicity, income, marital status, employment status, number of siblings - in order to make comparisons e.g. btw males and females group membership might be asked - e.g. illnesses/medical info may be asked to determine health and quality of life III. behaviors - past/intended future behaviors - e.g. how many times did you exercise last week - e.g. how many children will you have - e.g. have you ever been so depressed you called in sick to work b. Question Wording - biggest problem stem from difficulties with understanding the question including: (a) unfamiliar technical terms (b) vague or imprecise terms, (c) ungrammatical sentence structure (d) phrasing that overloads working memory (e) embedding the question with misleading info How to write a good question: I. unnecessary complexity: question should be simple II. double-barreled questions: avoid questions that ask 2 things at once III. loaded questions: are written to lead ppl to respond in one way, questions with emotionally charges words such as rape, waste, immoral, dangerous may influence the way that ppl respond and thus lead to biased conclusions IV. negative wording: e.g. do you believe that the city should not approve the proposed women’s shelter? This leads to reduction in scale reliability and validity V. “yea-saying” and “nay-saying”: when you ask several questions about a topic, there is a possibility that a respondent will employ a response set to agree or disagree with all the questions leading to yea-saying” and “nay-saying” response sets. The respondent may actually agree but they could also be agreeing with anything you ask, in order to avoid this is to word the questions so that consistent agreement is unlikely e.g. some questions are phrases so that agreement means the respondent is lonely and others with the meaning reversed so that disagreement indicates loneliness  Responses to Questions: What Kind of Data Are You Seeking? a. Closed-Versus Open-Ended Questions - closed ended questions (more structured): a limited number of response alternatives are given  most useful when the dimensions of the variables are well defined, favored from a quantitative approach - with open ended questions (costly): respondents are free to answer in any ways they like  most useful when the researcher needs to know what ppl are thinking and how they naturally view their world, used by both but favored from a qualitative approach b. Rating Scales for Closes-Ended Questions - rating scales ask ppl to provide “how much” judgments on any number of dimensions - they can have diff formats which depends on factors such as the topic being investigated - e.g. most direct scale presents ppl with 5 or seven response alternatives with labels defining either just the endpoints or all points I. labeling response alternatives - fully labels scales are more reliable than partially labeled scales - labels help to clearly define the meaning of each alternative, reducing error from relying on a persons idiosyncratic interpretations II. number of response alternatives - more alternatives allow ppl sufficient options to express themselves - sometimes the neutral option can cause problems for reliability/validity b/c respondents use it when they don’t know how to respond - to avoid this you should offer an I don’t know/not applicable option - another problem is deciding whether to have an odd/even number of alternatives - you could force participants to choose one side or the other by getting rid of the neither agree or disagree option and keeping a 6 point - sometimes a balanced scale is not possible/desirable this is done to force respondents to make finer distinctions - there are two types of labeling alternative: (1) high frequency scale: most alternatives indicate a high frequency of e.g. exercise (2) low frequency scale, you would probably use a high frequency scale if you were studying ppl who generally don’t exercise a great deal III. graphic rating scale - requires a mark along a continuous 100 mm line that is anchored with descriptions at each end - a ruler is placed on the line to obtain the score on a scale that ranges from 0-100 IV. semantic differential scale - a way to measure the meaning people ascribe to concepts - respondents rate any concept (ppl, objects, behaviors, ideas) on a series of bipolar adjectives using 7-point scales - e.g. smoking cigarettes: good ___ ___ ___ ___ ___ ___ ___ bad - anything can be measured using this technique, concepts rated using semantic differential scales are rated along 3 basic dimensions: 1. Evaluation: adjectives such as good-bad, wise foolish, kind-cruel, 2. Activity: active-passive, slow-fast, excitable-calm, 3. Potency: weak-strong, hard- soft, large-small V. non-verbal scale - instead of words or numbers e.g. use faces for young children who may not understand other types of scales  Finalizing the Questionnaire a. Formatting the Questionnaire - ask more interesting/important questions first - shorter questionnaires have higher response rates b. Refining Questions - give survey to group of ppl and have them “think aloud” while answering them - the “think aloud” procedure will allow you to see how ppl interpret each question and how they respond to the response alternatives  Administering Surveys a. Questionnaires - presented in written format and respondents write or type their answers - administered in person to groups or individuals through the mail, on the internet and with other technologies - positive features: 1. Less costly, 2. Anonymous, - Negative features: 1. Require respondents be able to read and understand questions (motivation) 2. Boring (attention) I. personal administration to groups or individuals - researcher is present to ask questions - captive audience that is likely to complete questionnaire once started II. mail surveys - ppl who do not live near, low response rate, no one is present to help III. internet surveys - open and closed ended questions can be administered - responses immediately sent to the researcher - can build databases of ppl willing to participate as well as forward to potential participants - can be easily advertised - problem: the true characteristics of the individuals providing info for the study are ambiguous IV. other technologies - e.g. instead of asking ppl to remember how they felt or acted over the past week, researchers use computerized experience sampling to contact participants at various times through the day on their cellphone and ask them to provide immediate reports of their current activities/emotional experiences b. Interviews - several implications because it involves an interaction btw ppl: - first: more likely to agree to answer questions than a mailed questionnaire - second: a rapport is developed that motivates person to answer all the questions - third: interviewer can clarify problems with understanding and ask follow-up questions - problem: interviewer bias which are all the biases that can arise from the fact that the interviewer could subtly bias the respondents answers by inadvertently showing approval/disapproval of certain answers - if there are several interviewers they could possess characteristics that may influence respondents answers - interviewers may have expectations that could lead them to see what they are looking for in respondents answers  biasing their interpretations of responses and leading them to probe further for an answer from certain respondents but not others I. face-to-face interviews (qualitative and quantitative) - expensive/time consuming, used when sample size is small/clear benefits II. telephone interviews (qualitative and quantitative) - large scale surveys, less expansive, data is collected quickly - computerized telephone survey techniques are even lower in cost by reducing labor and data- analysis costs III. focus group interviews (qualitative) - a focus group is an interview with a group of 6-10 individuals brought together for 2-3 hours - any topic can be explored, ppl are selected b/c they have interest/knowledge in topic - usually a sort of incentive to participate - questions are open-ended and asked of the whole group so group interaction is possible - group discussion is recorded/transcribed to find themes and areas of group consensus/disagreement (qualitative) or even content analyzed (quantitative)  Survey Designs to Study Changes Over Time - panel study (longitudinal design) is when the same ppl are tracked and surveyed at 2 or more points in time, particularly when measuring community members - two panel study ppl are surveyed at 2 points - they are particularly important when the research question addresses the relationship btw one variable at time one and another variable at some later time - studies that have a repeated measure or longitudinal component present researchers with a challenge e.g. societal/institutional changes might interfere with responses over time - another concern is that participants drop out of the study over time e.g. a sample that starts out random can become biased over time b/c of non random differences in the ppl who remain vs. leave the study  Sampling From a Population - population is a set of individuals of interest to the researcher - most research projects involve sampling participants from that pop - if a pop is large we can sample a selected sample from the pop of interest and use info obtained from the respondents to precisely estimate characteristics of the pop as a whole (statistical theory) a. Confidence Intervals - is a range of plausible values for the population value, values outside the confidence interval are implausible - the best estimate of the pop value is the sample value, but b/c you have only a sample not the entire pop your result may be in error (sampling error/margin of error)  when you study one sample the obtained result deviates from the true population value whenever there is sampling error - the confidence interval gives you info about the likely amount of the error b. Sample Size - larger sample size reduces measurement error and reduces the size of the confidence interval - larger sample are more likely to yield data that accurately reflect the true pop - sample size can be determined using a mathematical formula that takes into account the size of the confidence interval and the size of the pop you are studying - is NOT a constant percentage of the population size  Evaluating Samples - when the research has high external validity this means it can be generalized to the broader pop - this is done by ensuring that the sample used is unbiased by randomly sampling from a population that contains all individuals in the population and then you would contact and obtain completed responses from all individuals selected to be in the sample - even if random sampling is used there can be bias from 2 sources: sampling frame used and poor response rates st a. Sampling Frame (1 major source of bias) - the actual population of individuals or clusters from which a random sample will be drawn - how well does the sampling frame match the pop of interest, the severity of the bias impacts how much external validity a study has b. Response Rate (2 major source of bias) - a survey is the percentage of ppl in the sample who actually completed the survey - is 1 way of indicating how much bias there might be in the final sample of respondents, non respondents may differ from respondents in any number of ways, including age, income, marital status and education - the lower the response rate, the greater the likelihood that such biases may distort the findings, limiting the ability to generalize the findings to the population of interest (low external validity)  Sampling Techniques a. Probability Sampling I. simple random sampling II. stratified random sampling III. cluster sampling b. Non-probability Sampling I. convenience sampling II. purposive sampling III. quota sampling  Reasons For Using Convenience Samples Experimental Design and Conducting Experiments Chapter 8/9 Chapter 8 CONFOUNDING AND INTERNAL VALIDITY PLANNING A BASIC EXPERIMENT INDEPENDENT GROUPS DESIGN Pretest-posttest design Advantages and disadvantages of
More Less

Related notes for PSYB01H3

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit