Class Notes (835,873)
Canada (509,459)
Psychology (2,794)
PS295 (50)
Lecture

PS 295 Course notes

9 Pages
151 Views
Unlock Document

Department
Psychology
Course
PS295
Professor
Roger Buehler
Semester
Fall

Description
Criteria of a scientific approach: -systematic empiricism (observations, structure) -public verification(experiments can be replicated) -solvable problems (something that has an answer that is objective and can be found) Basic research: increases knowledge about a subject, understand something Applied research: finds a solution for a specific problem Psychological research attempts to: describe, predict, explain, sometimes control, behavior. Ways of learning: Authority (gaining knowledge from an expert in the subject) Consensus (common sense) Intuition (gut feeling) Rationalism: logical reasoning Experience Theories hypothesis prediction (general to specific) Theory: explain a large number of facts, findings, propositions that attempt to specify the interrelationships among a set of concepts Hypothesis: statement of relation thought to exist between 2 characteristics (ex. If-then), Seeks general principles that account for many observations. Has falsifiability (can be proven wrong), parsimony (no unnecessary assumptions, simplest explaination), strength of empirical support ( “methodological pluralism”- tested with several methods, “strong inference”- pitting two hypotheses against one another) Prediction: what specifically are we expecting? Induction: reasoning with specific instances to arrive at a general proposition Deduction: using a general proposition to explain a specific instance Types of research questions: Causal- changes in x result in changes of y Description/classification: characteristics of x Relationship/ association Primary sources: journal articles Secondary sources: review articles, meta-analysis Variable: anything that can take on different values across a set of objects Qualitative: type Quantitative: amount, measurable Discrete: no meanings in between (ex. Pregnant or not) Continuous: meaning in between, fractions (ex. Height) Hypothetical construct: variables that can’t be directly observed, such as social skills, memory, learning, attitude, happiness, intelligence Operational definition: precise method by which the variable is manipulated or measured- must be observable operation Conceptual: found in the dictionary Issues in manipulating variable: manipulate only the variable, have a number of levels (high, med, low), strength of manipulation (stronger may be more effective but less impressive), type of manipulation (environmental, instructional, invasive). Validity of manipulations: Construct validity: adequacy of operational definition Use multiple manipulations Manipulation check: see if you’ve actually manipulated the variable Measurement: assigning values meaningfully to get objects, behaviours, events, assign a number Self report, observational, physiological Scales: nominal (each one has a meaning, identity) Ordinal: also has meaning and order Interval: also equal intervals Ratio: also has true zero Inter-rater reliability - Have a code for the observers to follow that is structured to code for presence of specific behaviours, train the observers, - Standardize administration - Look to see how often they agree, percentage agreement, number of agreements over agreements+ disagreements. - Minimize coding errors - Reliability increases as increased observers, observations, occasions Face validity: if the experiment captures what it is supposed to capture (weakest way of assessing validity-subjective) Criterion related validity: is it related to a relevant behavioural criterion. Need to identify a specific behavior or outcome that directly represents the variable Predictive validity: future criterion Concurrent validity: current criterion Convergent validity: does it correlate with other measures it should correlate with Discriminant validity: does not correlate with measure it should not correlate with Range effects: when participants only/primarily respond in a certain range. Responses aren’t spread out. Floor effects are lower, ceiling is higher. Ex. Scores from 0-20 and responses are all 0-5 or 18-20. Means it may not be sensitive enough to see differences in data. Demand Characteristics: if subject knows how they are supposed to behave it may change their behavior EMG electromyography- indirect measure of attitudes, muscle movement in the face Stapel case: fabricated data uncovered by students and lab assistants Tri-council policy: respect for human dignity, 3 core principles ,respect for persons (people’s autonomy, free to act in ways that they choose), concern for welfare (quality of the experience of their life, not to affect their physiology, psychology, economy) justice ( inclusiveness and equitable distribution of rewards). Process for ethics approval of research with humans: Request for Ethics review form. Also an expedited version for minimal risk research by REB of psychology and small committee. Uses cost/benefit analysis. REB membership: faculty from various disciplines and community members. REB can approve, require changes, reject. REB, faculty research, graduate student research, undergraduate student research Debriefing required for deception studies. Sampling: how a researcher chooses individuals of a population to study Descriptive research: primary interest is characterizing typical or average response in a population Representative sample: closely reflects the characteristics of the broader population, to try and describe a larger population Sampling error- extent to which sample can be generalized to population Error of estimation (margin or error)- mathematical estimate of degree to which sample data differs from data of the entire population ex (+- 3%) Lower margin of error with a larger sample size. Greater population size leads to greater sampling error. Greater variability of data leads to greater sampling error. This is only used for probability sampling designs. Probability sampling: a sample for which the researcher knows the probability that any individual in the population is included in the sample. Requires the researcher “knows” the entire population and is time consuming and difficult and in the end does not entirely guarantee a high degree of representativeness. Epsem design: all cases in the population have an equal probability of being selected in the sample Probability sampling: Simple random sampling (example of epsem technique): one individual being selected has no bearing on the selection of another individual. Every possible sample has the same chance of being selected. Must have a clear list of population to create a table of random numbers to be entered into a computer program of random numbers. May not be representative, just by chance, especially with a small sample. Stratified random sampling: this is useful when you know there are subgroups within the population that could be important. First divide the population into subgroups, do simple random sampling for each subgroup. Sometimes an equal sample from each group is important and sometimes it is done proportionately to the population. Cluster sampling: divide population into groups “clusters”, select subset of clusters, randomly select from those clusters. Multi stage sampling process. Easier to contact participants in geographical groupings. Non response problem: if a certain type of person of the sample doesn’t respond or can’t be contacted it can affect the sample. They avoid it by following up many times, incentives, and also report response rate in articles so they can see if there will be a bias. Another method is by trying to see if respondents differ from non-respondents in some key way to determine if there will be a large bias. Non probability: Convenience: what is available ex. Psych students Quota: convenience with certain proportions ex half males half females Purposive: most informed people, people are chosen for a reason Descriptive research: goals, establish the typical response, describe a particular invididual, group or behavior. Techniques: survey research, demographic research, epidemiological, archival, case studies, observations. Typically not designed to test hypothesis, least often used, simply describes a certain population. Descriptive statistics: used to summarize characteristics of a population, purpose is to summarize and make clear a set of scores from a research study Frequency distribution: tally chart Grouped frequency distribution- create mutually exclusive intervals Relative frequency, how often it occurs Graphs: Graphical description, Histogram (bar), bar graph (categorical) Polygon (dots connected to
More Less

Related notes for PS295

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