Class Notes (835,828)
Canada (509,428)
Psychology (3,977)
PSYC 2360 (109)
Gonzales (13)
Lecture

Lecture Notes .docx

16 Pages
127 Views
Unlock Document

Department
Psychology
Course
PSYC 2360
Professor
Gonzales
Semester
Fall

Description
Chapter 9: Correlational Research Designs Scatterplot - graph in which the x axis indicates the scores of the predictor variable and the y axis represents the scores on the outcome variable - point is pointed for each individual at the intersection of their scores - y-axis  outcome variable - x-axis  predictor variable Regression Line - straight line of “best fit” drawn through the points on a scatterplot - line in which the squared distances of the points form the line are minimized Pattern of Relationships - linear relationship o when the association between the variables on the scatterplot can be easily approximated with a straight line - positive linear - negative linear - nonlinear relationships o not all relationships between variables can be well described with a straight line o independent  when there is no relationship at all between the two variables o curvilinear relationships  relationship that change in direction The Pearson Correlation Coefficient (r) - summarize and communicate the strength and direction of the association between two quantitative variables - frequently designed by the letter r - values range from r = - 1.00 to r = +1.00 - strength or effect size of the linear relationship o .10 – weak o .30 – moderate o .50 – strong Interpretation of r - significant r o linear association between the variables - coefficient of determination o indica2es the proportion of variance accounted for o is is r Factors That Distort r - curvilinear relationship between X and Y - restriction of range o cases in which the range of the variables is restricted - outliers o extreme values may have an extreme effect on r Chi-Square Statistic (X ) - Chi-square statistic o must be used to assess the relationship between two nominal variables o aka chi-square test of independence o indicates that there is an association between the two variables, but the specific pattern of the association is usually determined through inspection of the contingency table Multiple Regression - multiple regression o statistical analysis procedure using more than one predictor variable to predict a single outcome variable - regression analysis provides o multiple correlation coefficient (R)  ability of all of the predictor variables together to predict variables and the outcome variable  not exactly the same as zero-order correlations because they represent the effect of each predictor holding constant or controlling the effects of the other predictor Correlation and Causality - correlational research o cannot be used to draw conclusions about the causal relationships among the measured variables o although the researcher may believe the predictor variable is causing the outcome variable, the correlation between the two variables does not provide support for this hypothesis - CORRELATION DOES NOT EQUAL CAUSATION o directionality problem  reverse causation  causal direction is opposite what has been hypothesized  reciprocal causation  two variables cause each other o interpersonal interactions o third variable problem  could be related to both variables  correlation may not indicate a direct causal path but may arise because both variables are related to a third variable that isn’t being studied Correlational Data to Test Causal Models - one approach to ruling out reverse causation is to use a longitudinal research design o same individuals are measured more than one time and the time period between the measurements is long enough that changes in the variables of interest could occur When do we use correlation studies? - allows you to measure things you might not otherwise be able to - some variables we cannot manipulate o not ethical to assign participants to a smoking condition Chapter 10: Experimental Research: One-Way Designs Experiment - investigator manipulates the IV or variables by arranging difference experiences for the research participants and then assess the impact of these different experiences on one or more measured dependent variables o manipulation  experimenter systematically alters the levels of a variable o random assignment  every participant has an equal change of being assigned to any specific condition in an experiment Establishment of Causality - association o probabilistic - temporal priority o if event A occurs before event B, then A could be causing B o if event V occurs after event B, then A cannot be causing B - control of common – causal variables o eliminate confounds  variables that may produce spurious relationships o difficult unless true experiment  experimental manipulation - in experiments o covariation of cause and effect  measured by comparing treated and untreated groups o temporal priority  taken care of because you intervene before you measure outcome o no alternative explanations  taken care of through random assignment One-Way Experimental Design - one-way experimental design o has one IV o IV is created – manipulated - levels – experimental conditions o refers to the specific situations that are created within the manipulation Equivalence and Control - equivalence can be created through o between-participants designs  with different but equivalent participants in each level of the experiment  OR o repeated-measures designs  with the same people in each of the experimental conditions Random Assignment - random assignment to conditions o most common method of creating equivalence among the experimental conditions o do not confuse with random sampling o extraneous variables may be present but their average score will be present but their average score will be the same in each conditions, > N Variety and Number of Levels - experimental conditions o level of the IV in which the situation of interest was created - control condition o level of the IV in which the situation was not created - adding more levels o variables experimental conditions o two experimental, one control Detecting Nonlinear Relationships - if only two levels, it is difficult to draw conclusions about the pattern of relationship - 3 levels: higher N Repeated-Measures Designs - within-participants or within-subjects design o equivalence is created by having the same people participate in each condition o repeated-measure because the dependent measure is assessed more than one time for each person - one-way repeated-measures design – counterbalanced - advantages o statistical power  because the response of an individual in one condition can be directly compared to the same person’s response in another condition, the statistical power is greater than the power of a between- participants design in which different people are being compared across conditions - disadvantages o carryover  difficult to ensure that each measure of the dependent variable is being influenced only by the level it is designed to assess o practice and fatigue  if we the DV involves physical skill, the participant might improve on the task over time or might become fatigued o solutions  increased time period between measures  counterbalancing  arrange the order in which the conditions are experienced so each condition occurs equally often in each position  estimate the effect Counterbalancing - counterbalancing o arranging the order in which the conditions are experienced so each condition occurs equally often in each position o when conditions are more than 3, we should use a Latin square design to use a subset of all of the possible orders, but to ensure that each condition appears in each other Latin Square - label all conditions with a letter - if 4 conditions ABCD - if 6 conditions  ABCDEF Key to ANOVA - “method of quantifying how far apart sample means are from one another” - if sample means are further apart than we would expect by change if the Ho were true, reject Ho - if they are not further apart, then we do not reject Ho - analysis of variance o analyzes the variability among the conditions means Logic of ANOVA - Ho is true o groups are from the same population o BW group variance = WI group variance o if all the means are equivalent, then there should be no differences among them except those due to change - Ho is false o BW group variance will be substantially larger than WI group variance o condition means should not all be the same, and thus there will be significantly more variability (more differences) among them than would be expected by change - S (variance) a measure of the dispersion of the scores on a variable - ANOVA compares the variance of the means of the dependent variable o between the different levels  to the variance of individuals on the dependent variable o within each of the conditions  F = (between group variance)/(within group variance) - if the two variances are equal o F=1 - BW is larger than WI o F >1 - F>1 is significant o manipulation has influence the dependent measure because the influence of the manipulation across the levels is greater than the random fluctuation among individuals within the levels ANOVA - conceptual explanation - analyze how the variance in participant’s scores on the DV in the whole study is due to o treatment variance o random error - if there is a significant effect on the IV on the DV, then much more variance in the DV should be due to treatment than due to random error Variance due to Treatment - how much do the conditions means different from each other? o step 1  look at the data broken down by conditions o step 2  measure the degrees to which the group means vary from the rand mean  “Between Groups Variance” Variance due to Random Error - step 1 o look at the variance from the mean within each condition - step 2 o add this variance up from 3 conditions to get your variance due to error o “within groups variance” ANOVA - compare your F statistic with the critical value for F Effect Size - What about practical significance? o how large is the effect? - eta-squared o small = > . 09 o medium => .09 to .25 o large = > .25 Chapter 3: Ethics in Research What are ethics? - moral principles that govern a person’s or group’s behaviour - refers to the moral correctness of a specific conduct Reasons for Ethics Guidelines - human beings often behave different - scientist sometimes engage in practices that may be questioned on ethical grounds o welfare of research participants o honesty in conducting and reporting scientific research o giving credit for ideas and effort o how knowledge gained through research should be used Basic Ethical Principles - scientists assess the ethical principles of each and every research project and realize that may have to change or potentially even abandon certain research procedures o individual values o ethical codes – APA Why do we have an agreed-upon set of basic ethical principles? - protect participants from physical and psychological harm - provide freedom of choice and participating in the research - maintain awareness of the power of differentials between researcher and participant - honestly describe the nature and use of the research to participants History of Ethics - prior to WWII there was little concern for the treatment of humans in research o unethical treatment of prisoners in Nazi concentration camps o 1947  those responsible were tried for their crimes at Nuremburg o led to the creation of Nuremberg Code  10 standards for researchers/physicians Post-Nuremberg Code - abuses and exploitations of humans in research continued o Tuskegee Syphilis Study  1932-1972  physicians deliberately denied men treatment for syphilis and prevented treatment from other sources o Milgrim Studies  early 1960s  interested in obedience  interest stemmed from Nuremberg Trials  advertised for a study of “memory and learning”  criticism  study criticized for extreme psychological stress  psychiatrist who interviewed a fraction of subjects found no “injurious effects” Physical and Psychological Harm - participation in behavioural research may produce emotional reactions - reactions may have long-term, negative outcomes - potential for lasting impact o Milgram (1974) participants did not report long-term negative outcomes Providing Freedom of Choice - financial need - naturalistic observational studies - organizational settings - student’s research credits - potential gains o knowledge, benefit to society, explain purpose and results Robbers Cave Experiment - Sherif - phase 1 o randomly assigned to two groups  N=22 o arrived and housed in separate cabins  eagles vs. rattlers - phase 2 o intergroup conflict - phase 3 o establishing peace Tearoom Trade Study - Laud Humphreys – 1970 - investigated male homosexual activities in public places - misrepresented his identity and intent - privacy of subjects was infringed Informed Consent - potential participant - 1. provides her or his demographic info - 2. is given informed consent form explaining o procedure of the research o who is conducting it o how is the result will be used o what is going to happen during the research - 3. is informed of his or her rights, including the freedom to leave the project at any point - 4. is given the opportunity to ask questions - 5. signs the form indicating that he or she has read and understood the info Problems with Informed Consent - participants may change their behaviour o not “real” o not generalizable - selection bias o may change who would participate in study o not generalizable Awareness of Power Differentials - avoid abuses of power - respect participant’s privacy o confidentiality and anonymity - relationship based on trust and respect Honestly Describing the Research - deception o occurs whenever research participants are not completely and fully informed about the nature of the research project before participating o active vs. passive Deception - as been lost of deception in social psych o altruism, aggression, stereotyping - undermines informed consent - simply wrong to mislead people? - damage to psychology’s reputation? - use of deception decreasing, most studies do not use deception Alternatives - stimulation study o participants are fully informed about the nature of the research and asked to behave “as if” they were in a social setting of interest - problems o may not be applicable o may not reflect reality o may still produce unethical activity Stanford Prison Experiment Video Discussion - what are some of the ethical issues surrounding the experiment? - was the argument that “we need to know why people behaviour like this” sufficient enough to outweigh the costs? Debriefing - debriefing o occurs immediately after the research has ended o designed to explain the purposes and procedures of the research o important in research that involves deception  assess effectiveness of deception and alleviate potential impact on participants Debriefing Techniques - postexperimental interview o participants’ reactions are assessed o participants may be asked to express their thoughts about the research - suspicion check o determines if participants believed the experimental manipulation or guessed the research hypothesis: do you know it there is more to this experiment? - process debriefing (after stressful/embarrassing situations) o an active attempt to undo any changes that might have occurred (positive mood) o it is often impossible to entirely undo the effects of experimental manipulations - explanation of the purpose of the experiment – education function o research hypothesis o how it is tested o goals of the research o time to ask questions o “don’t tell” Milgram Study: Debrief - 83.7% - glad to have been in study - 15.1% - neutral - 1.3% - sorry - 80% - more studies should be done - 74% - said they learned something - psychiatrist who interviewed a fraction of subjects found no “injurious effects” Other Ethical Issues - animal research - should we do animal research? o this is both an ethical and political question o humane treatment - must adhere to Code of Ethics regarding animal treatment Ensuring That Research Is Ethical - ethics is determined through a cost-benefit analysis o if the potential costs of the research appear to outweigh any potential benefits, then the research should not proceed o current state of thinking within society - institutional review board (IRB) o determines whether proposed research meets required regulations o cost-benefit ratio - researcher’s own ethics (Europe Corrently and Honestly Reporting Research Results - science is based on truth, and scientists are expected to the truthful in all aspects of their research - scientific fraud o occurs when a scientist intentionally alters or fabricates data o must take steps to correct errors o must not plagiarize Chapter 11: Experimental Research: Factorial Designs Factorial Experimental Designs - factorial designs o more than one IV is manipulated - each IV is referred to as a factor - factorial designs have as many factors there are IVs Describing Factorial Designs - 2 x 2 - 2 x 3 - 2 x 2 x 2 Results - main effects o impact of one factor on the DV when controlling for all other factors in the experiment - to determine main effects you look at your marginal means o when means are combined across the levels of another factor, they are said to control for or collapse across the effects of the other factors Interaction - interaction o a pattern of means that may occur in a factorial design when the influence of one IV on the DV is different at different levels of another IV Simple Effect - effect of one factor within a level of another factor Conclusions - if only main effect(s) are significant, interpret them - if only the interaction is significant, interpret it - if the main effect(s) are significant and so is the interaction, interpret the main effect as qualified by the interaction How to Recognize Main Effects and Interactions - main effects o look at the table of means and graph - interactions o look at the graph Effect of Height and Gender on Yearly Income - IV 1: Height o tall vs. short - IV 2: Gender o male vs. female - fully crossed design o tall males, tall females, short males, short females - DV: yearly income Example 3-Way - DV: Cognitive performance o measured using ERP - IV 1: activity preference o morning lark OR evening owl - IV 2: time of day o morning OR evening - IV 3: food o no food OR food - null is that the two way interactions are the same at the different levels of the third variable Repeated Measures and Mixed Factorial Designs Means Comparisons - which group means are significantly different from each other? - pairwise comparison - planned or a priori comparisons o compare only the means involved in the hypotheses - post hoc comparisons o if not planned, this procedure takes into account the number of comparisons and controls for increase in experimental wise alpha  LSD, Tukey, Scheffe, Bonferroni Chapter 12: Experimental Control & Internal Validity Threats to the Validity of Research - threats of construct validity o does the manipulation of the IV appropriately create the conceptual variable of interest? - threats of statistical conclusion validity o type 1 and type 2 errors - threats to internal validity o can we trust the conclusions about the causal relationship between the IV and the DV - threats to external validity o can we generalize the results beyond the specific setting and participants to other places, people and times? Internal Validity - extent to which changes in the dependent variable can confidently be attributed to the effect of the IV, rather than the potential effects of confounding variables Experimental Control - occurs to the extent the experimenter is able to eliminate effects on the DV other than the effects of the IV o extraneous variables o confounding variables - extraneous variables o variables other than the independent variable that cause changes in the dependent variable  initial differences among the research participants
More Less

Related notes for PSYC 2360

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