Research In Psychology: Methods & Design PDF

# RESEARCH IN PSYCHOLOGY: METHODS & DESIGN - Eighth Edition ## Chapter 3. Developing Ideas for Research in Psychology ### Chapter Objectives - Distinguish between: - basic and applied research, - laboratory and field research, - qualitative and quantitative research. - Be able to form...

# RESEARCH IN PSYCHOLOGY: METHODS & DESIGN - Eighth Edition ## Chapter 3. Developing Ideas for Research in Psychology ### Chapter Objectives - Distinguish between: - basic and applied research, - laboratory and field research, - qualitative and quantitative research. - Be able to formulate a good empirical question. - Understand the need for operational definitions. - Describe research examples that have developed from everyday observations and serendipitous events. - Describe the defining features of a theory in psychology. - Describe how theories: - lead to empirical research, - are influenced by the outcomes of research, and - need to be productive, parsimonious, and testable. - Understand the importance of the "What's next?" question and the value of research that simultaneously replicates and extends prior research. - Distinguish between direct and conceptual replication and explain the importance of replication in psychological science. - Show how creative thinking occurs in science. - Use computerized databases (e.g., PsycINFO) to search for information about research in psychology. ## Varieties of Psychological Research ### The Goals: Basic versus Applied Research - **Basic** - designed to understand fundamental psychological phenomena. - example: stimulus factors affecting selective attention. - **Applied** - designed to shed light on the solution to real-world problems. - example: effect of cell phone use on driving. ### The Setting: Laboratory versus Field Research - **Laboratory** - greater control. - minimal mundane realism. - **Field** - more realistic - maximum mundane realism. - **Research Example 1** - Media violence, desensitization, and helping. ### The Data: Quantitative versus Qualitative Research - **Quantitative** - Includes quantitative data and statistical analysis. - **Qualitative** - Often includes narrative descriptions or interviews. - Some quantification of responses may occur. - Much research includes elements of both. ## Asking Empirical Questions - **Empirical questions** - Answerable with data. - Terms precisely defined. - **Operational definitions** - variables defined in terms of a clearly specified set of operations. - Hunger = 12 hours without food. - Frustration = consequence of being blocked from goal. - **Converging operations** - Understanding increases as studies with different operational definitions "converge" on the same result. ## Where to Research Ideas Come From? - **Our own observations** - sometimes from serendipitous events. - **Psychological theories**. - **Existing research** - development of research teams & the question of "what's next?". ## Developing Research from Observations of Behavior and Serendipity - **From observations** - Helping behavior (e.g., following Kitty Genovese murder). - **From serendipitous events** - Discovering something when looking for something else. - Skinner's first extinction curve (Chapter 1). - Box 3.1: Hubel and Wiesel (1959, 1962). ## Developing Research from Theory - **The nature of theory** - Summarizes, organizes, explains, provides basis for predictions. - Includes constructs: hypothetical factors involved in the attempt at explanation. - e.g., cognitive dissonance. - **The relationship between theory and research** - Hypotheses deduced from theory. - Outcomes/data provide or fail to provide inductive support for theory. - theories are never "true" nor "false". - **Attributes of good theories** - **Productivity** - good theories produce much research and advance our knowledge. - e.g., cognitive dissonance theory. - **Falsification** - good theories can be shown to be wrong (fail to be supported by the data). - Clever Hans example (Box 3.2). - Math capability ruled out (falsified). - **Parsimony** - good theories are concise and provide a simple explanation for results. - Clever Hans again. - Simpler (more "parsimonious") explanation (visual cues). - **Common misunderstandings about theories** - "It's not a fact; it's only a theory." - "It's just a theory; there's no proof." - "Here's my theory about that." ## Developing Research from Other Research - **Research teams and the "What's Next?" Question** - **Programs of research** - Series of interrelated studies. - **Research Example 2** - Retrieval practice and memory improvement. - **Replication** - **Direct replication** - An reproduction of the exact study procedures as the original study. - **Conceptual replication** - A partial replication, with new features added to extend the original study's findings. - **Ethics Box 3.3** - Questionable research practices and replication remedies. ## Creative Thinking in Science - **Pasteur**: "chance favors the prepared mind". - **Maze learning** - **prepared mind**: Small and Kline knowledgeable about animal behavior. - **chance**: Kline's comment about tunnels and Sanford's suggestion about Hampton Court (a) led to Small's redesign (b). - However, once an apparatus becomes established, creativity can be stifled. ## Reviewing the Literature - **Computerized database searches** - In psychology: PsycINFO. - Most recent info: www.apa.org/psycinfo - **Search tips** - Advanced search option (use of multiple search terms). - Using truncated search terms to avoid being too narrow. - Best strategy: trial and error. - **Search results** - Results list begins with most recent research. - Take note of source (e.g., journal article, book, dissertation). - you may limit your search by date or source too. - Read Abstracts provided when you click on the title. - **Table 3.2. Getting the most out of reading journal articles.** ## Summary - Researchers doing various types of research, depending on goals (basic vs. applied), setting (lab vs. field), and type of information collected (quantitative vs. qualitative). - Empirical questions are derived to attempt to better understand psychological phenomena. - Research ideas emerge from observations, theories, and/or pre-existing research. - Researchers become literate on a topic by exploring the research literature (e.g., via PsycINFO). - The research process has begun! # RESEARCH IN PSYCHOLOGY: METHODS & DESIGN - Eighth Edition ## Chapter 4. Sampling, Measurement and Hypothesis Testing ### Chapter Objectives - Distinguish between probability and nonprobability sampling. - Describe three varieties of probability sampling, and know when each is used. - Recognize the variety of behavioral measures used when conducting research in psychology. - Describe what psychologists mean by a construct and how measurable behaviors are developed and used to study constructs. - Explain how a behavioral measure is reliable and relatively free from measurement error. - Explain how a behavioral measure is valid, and distinguish several forms of validity. - Identify the defining features of nominal, ordinal, interval, and ratio scales of measurement, and know when each should be used. - Summarize data effectively using measures of central tendency (e.g., mean), variability (e.g., standard deviation), and visual displays (e.g., histograms). - Understand the logic of hypothesis testing and what is involved in making an inferential analysis of data. - Describe the criticisms of hypothesis testing and the suggested alternatives (e.g., confidence intervals). - Understand what is meant by (a) effect size and (b) the power of a statistical test, and know the factors that enhance power. ## Who to Measure - Sampling Procedures - **Samples vs. populations** - **Probability sampling** - **Random sampling** - Each member of population. has equal chance of being selected as member of sample. - Sometimes use a random number generator to select from population. - **Stratified sampling** - Proportions of important subgroups in population. are represented precisely in sample. - 75% female; 25% male (2 strata). - **Cluster sampling** - randomly select a cluster of individuals all having some feature in common. - campus survey: sample first-year students who live on-campus. - **Nonprobability sampling** - Does NOT provide representative samples, but are easier to do. - **Convenience sampling** - Select subjects who are available and convenient (e.g., Introductory Psychology "subject pool"). - Purposive sampling (e.g., Milgram non-use of university students). - **Quota sampling** - Similar to stratified sampling, but non-random. - **Snowball sampling** - Ask subjects to get their acquaintances to participate. - Often done with online surveys. ## What to Measure-Varieties of Behavior - **Developing measures from constructs** - Relates again to operational definitions. - **Research Example 3 – Habituation** - **Construct**: understanding of gravity. - **Measures**: preferential looking and time spent looking. - **Developing measures from constructs** - **Research Example 4 – Reaction Time** - **Construct**: visual imagery. - **Developing measures from constructs** - **Historical context – reaction time as a measure.** - **The "complication" experiment (Box 4.1).** ## Evaluating Measures - **Reliability** - Results from a minimum amount of measurement error. - Reliability = repeatability, consistency. - **Validity** - Measures what it is designed to measure. - Content validity (face validity). - Criterion (predictive and concurrent). - Construct (convergent and discriminant). - **Research Example 5 – Construct Validity** - Establishing construct validity of "Connectedness to Nature" scale. - Convergent: correlated with NEP scale and ecological behaviors. - Discriminant: did not correlate with SAT or social desirability. ## Scales of Measurement - assigning numbers to events, characteristics, or behaviors. - **4 Scales of Measurement:** 1. Nominal Scales 2. Ordinal Scales 3. Interval Scales 4. Ratio Scales - **Nominal Scales** - assign numbers to events to classify them into one group or another. - numbers are used as names [categorical]. - **How used:** 1. assign individuals to categories. 2. count the number of individuals falling into each category (reported as frequencies). - **Example:** - Verdict: 0 = not guilty, 1 = guilty. - **Ordinal Scales** - numbers are used to indicate rank order. - **How used:** 1. rank order (1st, 2nd, 3rd, etc.) individuals based on one or several other pieces of data. - **Example:** - Four students' class rank (based on GPA): 1, 2, 35, 100. - **Interval Scales** - scores indicate quantities. - equal intervals between scores. - score of zero: just a point on the continuum. - a score of zero does not indicate 'absence' of something. - **How used:** 1. calculate score from participants' responses on a test. - **Examples:** - temperature, IQ scores, scores from personality tests (see Box 4.2) - **Ratio Scales** - scores indicate quantities. - equal intervals between scores. - score of zero: does denote 'absence' of something. - **How used:** 1. calculate score from participants' responses on a test - **EXAMPLES:** - # words recalled, # errors made in maze learning task, time to make a response (reaction time). - **Why do we need to know which measurement scales is being used?** - Because it guides our decisions on which statistical tests are appropriate to use! ## Statistical Analysis - **Descriptive and inferential statistics** - **Descriptive statistics** - Describe the sample data. - Measures of central tendency. - What scores are at the center of a distribution. - Mean, median, mode. - With outliers: median better than mean. - Measures of variability. - How spread out or dispersed scores are in a distribution. - Range, standard deviation, variance, interquartile range. - With outliers: interquartile range better than standard deviation. - Visual displays of data. - Histograms from frequency distributions. - With graphs, carefully examine Y-axis to avoid being misled (see Box 4.3). - CNN example – beware the Y-axis. - **Inferential statistics** - Inferring general conclusions about the population from sample data. - Examples: t-tests, ANOVAs. - **Null Hypothesis Significance Testing (NHST)** - **Null hypothesis** - No relationship (“no difference") between variables in the population expected, given our sample. - **Alternative hypothesis** - A relationship ("a difference) between variables in population is expected, given our sample. - A researcher's predictions often specifies the direction of the relationship (e.g., a positive correlation between variables). - **2 possible outcomes** - **Reject null hypothesis (with some probability)** - Conclude you found a significant relationship between variables. - **Fail to reject the null hypothesis** - Conclude you found no significant relationship between variables. - **Because you are testing a sample and making inferences about the population, your statistical decisions have a probability of being wrong!** - **Possible errors** - **Type I**: reject null hypothesis, but be wrong. - **Type II**: fail to reject null hypothesis, but be wrong. - **Type I and Type II errors** - **Interpreting failures to reject null hypothesis** - Extreme caution. - May be useful if the outcome is replicated. - Example: questioning a claim for the effectiveness of some new therapy; useful if studies consistently show lack of effect of therapy. - Publication bias and the file drawer effect. - **Beyond Null Hypothesis Significance Testing** - **Effect size** - Emphasizes the size of difference between variables, not merely whether there is a difference or not. - Useful for meta-analysis. - **Confidence intervals** - Range within which population mean likely to be found. - **Power** - Chance of rejecting a false null hypothesis. - Sample size an important factor. ## Summary - Various sampling procedures are available to find a representative sample from the population. - Statistics is the language we use to communicate our findings to others in our field. - We use statistics to estimate the likelihood that the behaviors we observe are valid and occur in the way we think...and less likely due to chance. - Then, we have a better understanding of the psychological phenomena we are investigating!