Psychology 2103 Quantitative Research Methods Fall 2024 PDF

Psychology 2103 Quantitative Research Methods Fall 2024 1 Reminders and Announcements Midterm: October 25th ‾ 25 multiple choice questions, 5 short answer questions ‾ Review sessions on October 23rd (6-8 pm) and October 24th (5-7 pm) ‾ Go over list of important topics to study ‾ Do practice questions 2 Measurement Error Every measure we obtain consists of two elements: – “True” Score Hypothetical concept – Error 1. Bias 2. Random Error Observed score = True score + Error Sources of Measurement Error Experimenter – Person administering the measure Participant Observer/Scorer – Person doing the coding not necessarily the same as the person administering the measure Experimenter Error Random Error – Noise, temperature, time of day, etc. – Solution? Bias Error – Experimenter characteristics Experimenter Characteristics When a particular aspect of an experimenter affects how participants respond Can be physical characteristics, such as gender, age, ethnicity, etc. Or personality characteristics, such as friendliness, hostility, or anxiety Controlling for Experimenter Characteristics Use standardized methods – Train experimenters to follow a set standard when administering procedures – Standardize aspects of the experimenter as much as possible (appearance, attitude, etc.) Replication Experimenter Error Random Error – Noise, temperature, time of day, etc. – Solution? Bias Error – Experimenter characteristics – Experimenter expectancies Experimenter Expectancies When the expectations of the experimenter affects how the participant behaves AKA “Rosenthal” effects – e.g., Study of intellectual “bloomers” Controlling Experimenter Expectancy Effects More possibilities for control here… Standardization – Instructions scripted, recorded in advance, or presented via the computer Objectivity – Make coding schemes as objective as possible – Automated recording equipment Participant Error Random Participant Error – Carelessness – Distraction – Solutions? Participant Bias – Demand Characteristics – Good Participant Effect – Response Bias Demand Characteristics and The Good Participant Effect Demand Characteristics Features of an experiment that seem to inadvertently cause participants to act in a particular way Good Participant Effect Tendency for participants to behave as they perceive the researcher wants them to behave Controlling for Demand Characteristics Conduct double-blind research – Neither the experimenter nor the participant know which condition is being run Give false information to participants about the true nature of the experiment – i.e., deception Participant Error Random Participant Error – Carelessness – Distraction – Solutions? Participant Bias – Demand Characteristics – Good Participant Effect – Response Bias Response Bias Yea-sayers & Nay-sayers Response Set When the context affects the way a participant responds Can be a factor of the experimental setting or the questions that are asked Social desirability can also be an issue Controlling for Response Biases Yea/Nay-saying – Include both “agree” and “disagree” items – Randomize question presentation Response Sets – Careful review of questions/setting – Pilot testing Observer Error Random observer error – Carelessness – Distraction Observer/scorer bias – Confirmatory Bias See what we want to see Controlling for Observer Error Eliminate human observer – Use mechanical measure instead – Can help with random and bias errors Limit observer subjectivity – Focus on observable behaviour – Use standardized coding scheme Make observer “blind” – Unaware of experimental condition Validity Construct Validity – The extent to which your manipulation or measure actually represents the claimed construct e.g., Does your measure of extraversion actually measure obnoxiousness? – Construct Validity Criteria Reliability Content validity Convergent validity Discriminant or divergent validity Establishing Reliability The repeatability or consistency of the research – Test-retest reliability – Inter-rater/inter-observer reliability – Internal consistency Improving Internal Consistency Add items/questions – Random error balances out e.g., think of your likelihood of doing well on a multiple-choice test with: – 1 question? – 50 questions? Create better items/questions – reduce potential variability in interpretation Content Validity The extent to which a measure covers all aspects of a construct e.g., Measuring love Commitment Sexual Attraction Convergent Validity The extent to which a measure correlates with other indicators of the same construct – i.e., People scoring high on your measure should also score high on other measures of the same construct Assessing convergent validity: – Similar measures – Known comparison groups – Other indicators of construct Discriminant Validity The extent to which your measure is distinguishable from other constructs – Distinguishable from related constructs People scoring high on your measure should not score too high on measures of similar constructs Love Liking Love score - Liking score correlation won’t be 0, but shouldn’t be.90 either. Discriminant Validity The extent to which your measure is distinguishable from other constructs – Distinguishable from unrelated constructs People who score high on your measure should NOT also score high on a measure of the “wrong” construct Social Love Desirability Love score - Social Desirability score correlation should be near 0 Beyond Reliability and Validity Reliability and validity not enough – Best measure = best fit for the research context Additional issues: – Is the scale appropriate to the context? e.g., How tall was the criminal? – Is the measure sensitive enough? Sensitivity Ability of measures to detect effects 1.Is the measure “strong” enough for what you want to study? – e.g., Looking for microbes with magnifying glass 2.Does your measure minimize the influence of error? – Is the “signal” clear relative to measurement “static” (i.e., error)? Achieving Sensitivity in Measurement Use measures with score variance: – Avoid restriction of range Ceiling and floor effects – Avoid all or nothing measures Ask how much instead – Add scale points to a rating scale – Pilot test measure Psychology 2103 Quantitative Research Methods Fall 2024 29 Reminders and Announcements Midterm: October 25th ‾ 25 multiple choice questions, 5 short answer questions ‾ Email today on times for October 23rd and October 24th ‾ Go over list of important topics to study Method Draft due on October 30th on D2L. 30 Experimental Research Designs Experimental Research A true experimental design: 1. aims at establishing a causal relationship between two variables – Do changes in the independent variable (AKA the manipulating/predictor variable) cause changes in the dependent variable (AKA outcome/responding variable) 2. Controls for/minimizes variability caused by extraneous variables Independent Variables Variable that is manipulated to determine its effects 4 types: – Physiological – Experience – Stimulus/Environmental – Participant Physiological IVs Manipulation of a participant’s normal biological state Examples… – How effective is a particular drug? – How does alcohol use affect people’s reaction time? Physiological IV Example Eight male participants were given either a) placebo or b) gradually increasing doses of testosterone (150, 300, & 600 mg/week each for 2 weeks) using a double-blind, randomized design. Participants were tested both before and after the series of injections. During the experimental session, participants could press a button to accumulate points exchangeable for money (non-aggressive response) or press another button to subtract points from a fictitious opponent (aggressive response). Aggressive responding was instigated by subtracting points from the participant which was attributable to the fictitious opponent. Testosterone administration resulted in a significantly Experience IVs Manipulation in the amount or type of training or learning Examples… – Testing a school program for teaching children abduction-prevention skills Stimulus/Environmental IVs Manipulation of some aspect of the environment Examples… – Examining the effects of changing various conditions on worker performance – Looking at whether the presence of a weapon decreases ID accuracy Participant/Subject IVs Aspects of the participant treated as if they are IVs Some examples… – Age, gender, personality traits, etc. Overview of the Study Design Road Crossing Young Simulation 30-45 yrs Cross or Young-Old Don’t 60-69 yrs Cross? Old-Old Speed & Distance > 75 yrs of car varied Participant/Subject IV Example The possibility that overrepresentation by older pedestrians in serious injury and fatal crashes is due, in part, to age-related diminished ability to select gaps in oncoming traffic for safe road- crossing was examined using a simulated road- crossing environment. Three groups of participants were tested, young (30-45 yrs), young-old (60-69 yrs), and old-old (>75 yrs). The results showed that despite an apparent ability to process the distance and speed of oncoming traffic when given enough time to do so, significantly more old-old adults were willing to cross the road when there was an insufficiently Dependent Variables Variable that is measured to see the effects of the independent variable. How to select your DV – Should relate back to your hypothesis – Must be operationally definable – Should be both VALID and RELIABLE Types of DVs Correctness – How many were right? Rate/Frequency – How often did it occur within a certain amount of time? Degree or Amount – How much of it was there? Latency or Duration – How fast did it happen? How long did it last? Types of Extraneous Variables: Nuisance Variables Unwanted variables that increase the variability of scores within all groups Affects ALL groups… Makes it harder to see the effect Types of Extraneous Variables: Confounding Variables Unintended variables that create a systematic difference between the groups on the DV Render findings meaningless Spot the Confounder! Example 1 Michelle is a hard-working graduate student. In addition to studying for her classes, she conducts research on human memory. Her current project is quite complex and involves the testing of four different groups of participants. To accommodate her busy schedule, Michelle decides to test two groups of participants early in the morning and the remaining two How to be a “Control Freak” Step 1: – Randomization Each participant has an equal chance of being assigned to any group in an experiment – Not possible to know all the potential extraneous variables that could be at play – Randomization is our best chance of “squaring the playing field” (aka making sure there are no preexisting differences between groups) How to be a “Control Freak” Step 2: – Elimination Specific extraneous variables completely removed from the experiment – Must know what the potential extraneous variables are – Not always easy in practice How to be a “Control Freak” Step 3: – Constancy When it is impossible to completely remove an extraneous variable, a researcher may try to minimize its effects by having it remain constant for all participants – Often refers to testing conditions… e.g., noise, temperature, lighting, time of day, etc. How to be a “Control Freak” Step 4: – Balancing When it is impossible to completely remove an extraneous variable, a researcher may try to minimize its effects by distributing it to all groups equally – Examples of extraneous variables where balancing can help… All experimenters test ALL conditions equally Sequence or order effects & carryover effects What are order and carryover effects? Order Effects – When position in a series affects how participants respond – Doesn’t depend on the EVENT but rather the POSITION Carryover Effects – When the effects of one event influence responses to the next event – Depends on the EVENT not the POSITION How can balancing help? i.e., what specific balancing strategy? Counterbalancing – Vary the order in which items are presented – Can be either: Within-subject or within-group Complete or incomplete Complete Counterbalancing To truly be complete, there are 3 criteria that must be satisfied… 1. Each event must be presented to each participant an equal number of times 2. Each event must occur an equal number of times at each session 3. Each event must precede and follow Psychology 2103 Quantitative Research Methods Fall 2024 53 Reminders and Announcements Midterm: October 25th ‾ 25 multiple choice questions, 5 short answer questions ‾ Do the practice questions posted on D2L. ‾ More will be uploaded this week and the next as we approach the midterm date ‾ We will go over answers during review sessions on October 23rd and October 24th ‾ Go over list of important topics to study 54 Interpreting Standard Deviation: Empirical Rule (68 – 95 – 99.7 Rule) For data with a (symmetric) bell- shaped distribution, the standard deviation has the following characteristics: About 68% of the data lie within 1 SD of the mean About 95% of the data lie within 2 SDs of the mean About 99.7% of the data lie within 3 SDs of the mean 55 Interpreting Standard Deviation: Empirical Rule (68 – 95 – 99.7 Rule) 99.7% within 3 SDs 95% within 2 SDs 68% within 1 SD 34 34 % % 2.35 13.5% 13.5% 2.35 % % x  3s x  2s x s x x s x  2s x  3s 56 Example: Using the Empirical Rule In a survey conducted by the Stats Canada, the sample mean height of women in Canada (ages 20-29) was 64 inches, with a sample standard deviation of 2.71 inches Estimate the percent of the women whose heights are between 64 inches and 69.42 inches 57 Solution: Using the Empirical Rule Because the distribution is bell-shaped, you can use the Empirical Rule… 34 % 13.5 % 55.87 58.5 61.2 64 66.71 69.42 72.1 x  3s 8x  2s 9x  s x x s x  2s 3x  3s 58 Non-Experimental Methods I: Observational Research Descriptive Research Methods Using Census data to determine whether lower income families are more likely to have health problems Recording the behaviour of children on the playground to determine the prevalence of bullying Detailed observations of the abilities and behaviour of a man with synesthesia Jane Goodall’s research of chimpanzees in the wild What do these studies have in common? Types of Descriptive Studies Archival research Case studies Observational techniques – Naturalistic Observation – Participant Observation – Clinical Perspective Archival Studies Using data that has been previously recorded e.g., Does eye colour predict alcohol use? From archival data consisting of a sample of 10,860 Caucasian male prison inmates and 1,862 Caucasian women respondents in a national survey, found that people with light eyes consumed significantly more alcohol than those with dark eyes (Bassett & Dabbs, 2001). Archival Studies Issues? – May not have knowledge of characteristics of your sample Limits generalization – May have missing data values – As data often collected for a different purpose, may not be ideally suited to your research question – Cannot show cause-and-effect Case Studies Collecting detailed information about the behaviour of a single individual – Typically used to: Study one person over an extended period of time Take advantage of a rare/unusual occurrence – e.g., the wild boy of Aveyron Boy who apparently grew up in the wild, first found in 1797, in Southern France Critical period for language acquisition Case Studies Collecting detailed information about the behaviour of a single individual – Typically used to: Study one person over an extended period of time Take advantage of a rare/unusual occurrence – Issues? Generalization based on N of 1 Precludes cause-and-effect statements Observational Techniques Naturalistic Observation – Observing behaviour in the real world unobtrusively – e.g., What factors affect who cares for kids in public? Researchers stationed themselves in positions where they could observe pedestrian traffic, watched for families, & coded the gender of the primary caregiver, the gender of the child, the estimated age of the child, and the ethnicity of the family (Amato, 1989) Found men were more likely to care for kids Observational Techniques Participant Observation – When the observer imbeds him/herself within the group being studied – e.g., How do the homeless maintain a positive self-ID? Researcher worked as a volunteer at a homeless shelter and through discussions with the homeless people that stayed there developed a model that describes the progression by which people come to identify themselves as “homeless” and still maintain a positive self-identity (Farrington & Robinson, 1999). Observational Techniques Clinical Perspective – Descriptive approach aimed at understanding and correcting a particular behavioural problem – e.g., Does an “after-death” phone call help family members of a person with mental illness? Noted increased feelings of validation, comfort, and thankfulness to mental health provider for the call. Based on these findings, the authors recommended that after-death calls should be Observational Techniques Clinical Perspective – Descriptive approach aimed at understanding and correcting a particular behavioural problem Similar to but different from participant observation: 1.Client chooses clinician, whereas participant observer chooses others to study 2.Clinicians cannot be unobtrusive or passive because they have been asked to participate in the situation Observational Techniques Issues? – Reactivity When knowledge one is being observed affects his/her behaviour AKA “The Hawthorne Effect” – High on external validity, low on internal validity – Objectivity – Cannot make cause-and-effect statements Surveys & Questionnaires Descriptive Surveys Seek to determine what % of the population have particular characteristics, beliefs, or behaviours – Characteristics: Who are you? Do you live alone? How many children do you have? Do you suffer from depression? Diabetes? Asthma? Do you have a driver’s license? Are you employed? Surveys & Questionnaires Descriptive Surveys Seek to determine what % of the population have particular characteristics, beliefs, or behaviours – Beliefs: What do you think? Do you believe in gay marriage? Which do you like better? Pepsi or Coke? Who do you think will win the Stanley Cup? How would you rate the service at this restaurant? Surveys & Questionnaires Descriptive Surveys Seek to determine what % of the population have particular characteristics, beliefs, or behaviours – Behaviours: How will/do you act? Who do you plan to vote for? Do you engage in unprotected sex? Do you smoke? Would you buy this product? What television programs do you watch? Surveys & Questionnaires Descriptive Surveys Seek to determine what % of the population have particular characteristics, beliefs, or behaviours Using a sample to draw conclusions about the population Goal – Want it to be representative of the population Failure to represent the people… An example 1936 Presidential Election – Franklin D. Roosevelt versus Alf Landon Literary Digest Poll – Ballots mailed out to residential telephone subscribers and automobile owners – 2.3 million people responded What went wrong? Sampling bias – Roosevelt had more support from the working class because of his attempts to mitigate the effects of the depression Self-selection bias – Roosevelt was the incumbent so those who were unhappy with him were more likely to respond These biases decrease representativeness of the sample, thereby compromising generalizability Surveys & Questionnaires Analytic Surveys Seek to determine the relevant variables and how they are related e.g., Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Step 1. What are the relevant variables? Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Aggression and health behaviours Need to find a way to assess these variables… Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Construct 1: Aggression – Made up of anger, verbal aggression, and physical aggression – The Aggression Questionnaire (Buss & Perry, 1992) 25 items composed of separate components meant to tap anger, verbal aggression, physical aggression, & hostility Uses a 5-point likert scale Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Anger Some of my friends think I’m a hothead. Once in awhile, I can’t control the urge to strike another person. I flare up quickly but get over it quickly. I have trouble controlling my temper. When frustrated, I let my irritation show. I sometimes feel like a powder keg ready to explode. I am an even-tempered person. Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Verbal Aggression I tell my friends openly when I disagree with them. I can’t help getting into arguments when people disagree with me. When people annoy me, I may tell them what I think of them. I often find myself disagreeing with people. My friends say that I’m somewhat Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Physical Aggression If I have to resort to violence to protect my rights, I will. I have become so mad that I have broken things. Once in awhile, I can’t control the urge to strike another person. I have threatened people I know. I can think of no good reason for ever hitting a person. Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Construct 2: Health Behaviours How many times in the last 3 months have you smoked cigarettes? – 1 = not at all – 2 = a few times every month – 3 = several times a week – 4 = regularly, 1 – 2 a day – 5 = regularly, 3 – 10 a day – 6 = regularly, 11 – 20 a day – 7 = more than 20 a day Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Construct 2: Health Behaviours How many times in the last 3 months have you drank alcohol? – 1 = never – 2 = once or twice – 3 = a few times – 4 = more than a few times, up to once a week – 5 = regularly, at least two times a week Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Construct 2: Health Behaviours How many times in the last 3 months have you paid attention to what you have eaten (i.e., tried to eat a healthy diet)? – 1 = not at all – 2 = a little – 3 = about half the time – 4 = most of the time – 5 = always Surveys & Questionnaires Analytic Surveys Seek to determine the relevant variables and how they are related e.g., Is aggression related to health behaviours in adolescents? (Piko et al., 2005) Step 1. What are the relevant variables? Step 2. How are they related? Is aggression related to health behaviours in adolescents? Results (Piko et al., 2005) Choosing your Survey Lots of surveys exist out there – Useful because have generally been tested However, what if no survey exists to measure your construct? How to develop a Survey 1. How will you administer your survey? Pros - Can be completed without researcher - Can be sent to a large number of people - With online surveys, little/no data entry - Random-digit dialing ↑random sampling - Can enter data immediately on computer -- Can clear up ambiguous answers Increased response rate -- Can control order that questions are Can clear up ambiguous answers - answered Can control the order questions are How to develop a Survey 1. How will you administer your survey? Cons - Can’t be sure who answered the survey - Can’t be sure questions answered in - order Caller ID and voicemail - Low response rates - Low response rates - Can’t use visual aids or measure nonverbal cues - Takes more Difficult to time of experimenter establish rapport & participant - Is more expensive - Possibility of interviewer bias How to develop a Survey 1. How will you administer your survey? 2. What kind of questions will you use? Yes-No: answer “yes” or “no” Forced Alternative: select between two alternative responses Multiple Choice: select the best response from several alternatives Likert: select a response based on a designated scale How to develop a Survey 1. How will you administer your survey? 2. What kind of questions will you use? 3. Write the items Questions should: Be short, clear, concise Use familiar vocabulary Be at an appropriate reading level for your sample Be specific Use positive and negatively phrased questions How to develop a Survey 1. How will you administer your survey? 2. What kind of questions will you use? 3. Write the items 4. Pilot-test! Preliminary test to try out procedures and make any needed changes or adjustments How to develop a Survey 1. How will you administer your survey? 2. What kind of questions will you use? 3. Write the items 4. Pilot-test! 5. Any other info you want to collect? 6. Create the survey instructions Tests & Inventories How do they differ from surveys? Surveys/Questionnaires – Examine an opinion on an issue or topic Tests/Inventories – Assess a specific attribute, characteristic, or ability of the person being tested Research Strategies Single-strata approach – Select from one subgroup of a population Cross-sectional research – Compares multiple subgroups at the same time Longitudinal research – Looks at one group over an extended period of time – This group is called a COHORT Group of people born during the same time period Psychology 2103 Quantitative Research Methods Fall 2024 97 Reminders Intro and References draft due on October 11th by 11:59 PM. – Only one submission per group – Refer to D2L for rubric and APA resources – Please submit group sheet as a separate document with all group member names – Must include three references other than MCTQ and EPI references which you can copy and paste from D2L. No class on October 11th for this course Midterm: October 25th 98 Last Class Measures of central tendency – Mean: population vs sample mean – Median, mode – Use depends on type of data you have – For a normal distribution, mean=median=mode Measures of variability – Range – Quartiles and IQR 99 For normal data, where is… The Mean? The Median? The Mode? 55.87 58.5 61.2 64 66.71 69.42 72.1 8 9 3 Height in inches of women in North America (ages 20- 29) 100 Appropriate use with scales of measurement Measurement Best indicator Scale of the ‘middle’ Nominal Mode (e.g., gender of participant) Ordinal Median (e.g., class standing based on grades) Interval Mean… if (e.g., temperatures) symmetrical Median… if skewed 101 Ratio Mean… if Measures of Variability: Interquartile Range Interquartile Range (IQR) Measure of distance between the first and third quartiles – Special kind of range that includes just the middle 50% of values – Benefits: Less affected by extreme values Also helpful for identifying outliers 102 Measures of Variability: Interquartile Range Quartiles – Positions in a range of values representing multiples of 25% Median = 2nd quartile 50% of scores fall below median (Q2); 50% scores above – First quartile (Q1) 25% of scores fall below Q1; 75% above – Third quartile (Q3) 75% of scores fall below Q3; 25% above 103 Example: Calculating the IQR 1. Find the median (2nd quartile) location – 22 26 33 34 36 36 38 – There are seven entries (an odd number), the median is the middle, or fourth, data entry Or the position is (7 + 1)/2 = position 4  If median location is a fraction, round down to nearest whole number 104 Example: Calculating the IQR 1. Find the median (2nd quartile) location – Position 4 2. Find quartile location Quartile location = (median location + 1)/2 = (4 + 1)/2 = 2.5 105 Example: Calculating the IQR 1. Find the median (2nd quartile) location – Position 4 2. Find quartile location = 2.5 3. Find 1st and 3rd quartiles Q1 location = Position 2.5 from lowest score Q3 location = Position 2.5 from highest score 22 26 33 34 36 36 38 (26 + 33)/2 = 29.5 (36 + 36)/2 = 36 106 Example: Calculating the IQR 1. Find the median (2nd quartile) location – Position 4 2. Find quartile location = 2.5 3. Find 1st and 3rd quartiles = 29.5 and 36 4. IQR = distance between Q1 and Q3 (Q3 – Q1) = 36 – range The interquartile 29.5 of PCL scores for the 7 prisoners = 6.5 is 6.5. 107 IQR and Outliers Outliers – Extreme values that don’t “fit” with the rest of the data Rule of thumb for finding outliers using IQR – Scores > Q3 + (1.5 x IQR) are high end outliers e.g., 36 + (1.5 x 6.5) = 45.75 22– Scores 26 33 < Q1 34- (1.5 36 X 36 IQR)38 are low end outliers 19.75 45.75 108 Graphing IQR: Boxplots Whisker = largest Aka “box and value within Upper inner fence = whisker plots” upper Q3 + (1.5 x IQR) inner Improves on IQR fence Upper hinge = Q3 in H-spread = 3 ways: IQR 1. Includes all data Lower hinge = Q1 2. Clearly identifies Whisker = outliers smallest 3.Can plot side-by- Lower inner fence = value within Q1 - (1.5 x IQR) side boxplots to lower inner fence compare groups 109 Example: Drawing Boxplots 110 Measures of Variability: Deviation Deviation The difference between each score and the mean of the data set Population data set: – Deviation of xi = xi – μ Sample data set: – Deviation of xi = xi – x 111 Example: Finding the Deviation What is the deviation of each score? 38 33 22 36 34 36 26 112 Solution: Finding the Deviation 38 33 22 36 34 36 26 1. First, find the mean score (To find the mean score, divide the sum of the scores by the number of scores in the sample): Σx = 38 + 33 + 22 + 36 + 34 + 36 + 26 = 225 x 3695 225 x  527.9 32.14 n 7 113 Solution: Finding the Deviation 2. Determine the PCL Score (x) Deviation: x – x deviation for 38 38 – 32.14 = 5.86 each data entry 33 33 – 32.14 = 0.86 22 22 – 32.14 = -10.14 36 36 – 32.14 = 3.86 34 34 – 32.14 = 1.86 36 36 – 32.14 = 3.86 26 26 – 32.14 = -6.14 Σx = 225 Σ(x – x)= 0 Deviation scores always sum to zero 114 Measures of Variability: Deviation Deviation The difference between each score and the mean of the data set How is the deviation different from the IQR and boxplots? 115 Measures of Variability: Variance Variance Single number representing the average amount of variation in a set of scores ( x  x ) 2 Sum of Sample variance: 2 s  square n 1 s, SSx 2 ( x   ) 2   Population variance: N 116 Steps for Computing the Sample Variance In Words In 1. Symbols Find the mean of the x x sample data set. n 2. Find deviation of each x x entry. 3. Square each ( x  x )2 deviation. 4. Add to get the sum of SS x ( x  x ) 2 squares. 2  ( x  x ) 5. Divide by n – 1 to get s2  n 1 117 the sample Example: Finding the Sample Variance What is the sample variance? 38 33 22 36 34 36 26 118 Solution: Finding the Sample Variance 2 1. Determine Sum of Squares (SSx) 2 ( x  x ) s  n=7 n 1 PCL Score (x) Deviation: x – x SSx: Σ(x – x)2 38 38 – 32.14 = 5.86 (5.86)2 = 34.34 33 33 – 32.14 = 0.86 (0.86)2 = 0.74 22 22 – 32.14 = -10.14 (-10.14)2 = 102.82 36 36 – 32.14 = 3.86 (3.86)2 = 14.90 34 34 – 32.14 = 1.86 (1.86)2 = 3.46 36 36 – 32.14 = 3.86 (3.86)2 = 14.90 26 26 – 32.14 = -6.14 (-6.14)2 = 37.70 119 Σx = 225 Σ(x – x)= 0 SSx = Solution: Finding the Sample Variance 2. Divide the SSx by the sample size minus Sum 1 of squares, SSx 2 2  ( x  x ) 88.5 208.86 s   9.8 34.81 n 1 10 7 - 11 The sample variance is about 34.81. 120 Measures of Variability: Variance Variance Single number representing the average amount of variation in a set of scores ( x  x ) 2 Sum of Sample variance: 2 s  square n 1 s, SSx 2 ( x   ) 2   Population variance: N Typically, not reported in favour of reporting the STANDARD DEVIATION 121 Measures of Variability: The Standard Deviation Standard deviation Measure of the spread of scores out from the mean of the sample Most common way of describing the spread of a group of scores Steps for computing the standard deviation: 1. Calculate the variance 122 2. Find the square root Measures of Variability: The Standard Deviation Standard deviation Measure of the spread of scores out from the mean of the sample Most common way of describing the spread of a group of scores 2 Sample SD:  ( x  x ) s  s2  n 1 2 2  ( x   ) Population SD:     N 123 Steps for Computing the Standard Deviation In Words In Symbols 6. Find the square root ( x  x ) 2 of the variance to get s the sample n 1 standard deviation. 124 Solution: Finding the Sample Standard Deviation 1.Calculate the sample variance: 2 2 ( x  x ) 88.5 208.86 s   9.8 34.81 n 1 1077--111 2.Take the square root: 2 88.5 208.86 s s  3.1 5.9 96 125 Interpreting Standard Deviation Standard deviation is a measure of the typical amount an entry deviates from the mean – Thus, the more the entries are spread out, the greater the standard deviation 126 Interpreting Standard Deviation: Empirical Rule (68 – 95 – 99.7 Rule) For data with a (symmetric) bell- shaped distribution, the standard deviation has the following characteristics: About 68% of the data lie within 1 SD of the mean About 95% of the data lie within 2 SDs of the mean About 99.7% of the data lie within 3 SDs of the mean 127 Interpreting Standard Deviation: Empirical Rule (68 – 95 – 99.7 Rule) 99.7% within 3 SDs 95% within 2 SDs 68% within 1 SD 34 34 % % 2.35 13.5% 13.5% 2.35 % % x  3s x  2s x s x x s x  2s x  3s 128 Example: Using the Empirical Rule In a survey conducted by the Stats Canada, the sample mean height of women in Canada (ages 20-29) was 64 inches, with a sample standard deviation of 2.71 inches Estimate the percent of the women whose heights are between 64 inches and 69.42 inches 129 Solution: Using the Empirical Rule Because the distribution is bell-shaped, you can use the Empirical Rule… 34 % 13.5 % 55.87 58.5 61.2 64 66.71 69.42 72.1 x  3s 8x  2s 9x  s x x s x  2s 3x  3s 130 131 Psychology 2103 Quantitative Research Methods Fall 2024 132 Reminders Intro and References draft due on October 11th by 11:59 PM. – Only one submission per group – Refer to D2L for rubric and APA resources – Please submit group sheet as a separate document with all group member names No class on October 11th for this course Midterm: October 25th 133 Tabulating and Plotting Data 134 Tabulating Data The first quantitative step: – We need to find a way to summarize our numbers! How might we do that? 1. Create a frequency distribution 2. Create a graph Allows us to see our data pictorially Often preferable to tables 135 What is a Frequency Distribution? Frequency Distribution Class Frequency A table that shows Class 1–5 5 classes (intervals of width 6 – data) with a count of the 1 = 5 6 – 10 8 number of entries in each 11 – 6 interval 15 16 – 8 20 The frequency of a class 21 – 5 is the number of data Lower class25 Upper class entries in that interval limits 26 – limits4 30 136 Graphs of Frequency Distributions Frequency Histogram A bar graph that represents the frequency distribution – Horizontal (x) axis is quantitative and measures the data values – Vertical (y) axis measures the frequencies of the classes Consecutive bars must touch frequen cy data 137 Example: Constructing a Frequency Histogram for Internet Usage Let’s construct a frequency histogram for the Internet usage frequency distribution. Frequency Class 7 – 18 6 19 – 30 10 31 – 42 13 43 – 54 8 55 – 66 5 67 – 78 6 79 – 90 2 138 Example: Constructing a Frequency Histogram for Internet Usage 1. Calculate Class boundaries – The numbers that separate classes without gaps The distance from the Class upper limit of the first Class Boundari Frequenc es y class to the lower limit of 6.5 – 18.5 7– 6 the second class is 19 – 18 18 = 1. 19 – 10 Half this distance is 0.5. 30 First class lower boundary = 731 – 0.5 – = 6.5 13 First class upper boundary = 1842+ 0.5 = 18.5 139 Example: Constructing a Frequency Histogram for Internet Usage Class Class boundaries Frequenc y 7 – 18 6.5 – 18.5 6 19 – 30 18.5 – 30.5 10 31 – 42 30.5 – 42.5 13 43 – 54 42.5 – 54.5 8 55 – 66 54.5 – 66.5 5 67 – 78 66.5 – 78.5 6 79 – 90 78.5 – 90.5 2 140 Example: Constructing a Frequency Histogram for Internet Usage (Using Class Boundaries) 6.5 18.5 30.5 42.5 54.5 66.5 78.5 90.5 You can see that more than half of the subscribers spent between 19 and 54 minutes on the Internet during their most recent session 141 Example: Constructing a Frequency Histogram for Internet Usage (Using Midpoints) (Lower class limit)  (Upper class limit) Midpoint of a class: 2 142 Histogram vs. Bar Graph Histogram Used to show frequency 90 distribution of 80 responses 70 60 X-axis Frequency 50 - Continuous DV 40 - Bars are joined30 Y-axis 20 10 - Number of 0 participants 7 12 17 22 27 32 within each Multiple Choice Raw Scores class 143 Histogram vs. Bar Graph Bar Graph Full-time Undergraduate Enrollment Used to plot data 7000 Student Numbers 6000 X axis 5000 4000 - Categorical IV - Nominal Scale 3000 - Bars are separated2000 1000 Y axis 0 UNB-Fred UNB-SJ St. Thoma - Continuous DV NB University 144 Graphing Correlational Data Each entry in one data set corresponds to one entry in a second data set Graph using a scatter ploty – The ordered pairs are graphed as points in a coordinate plane – Used to show the relationship between two quantitative variables x 145 Example: Interpreting a Scatter Plot As petal length increases, what tends to happen to the petal width? Each point in the scatter plot represents the petal length and petal width of one flower From the scatter plot, you can see that as the petal length increases, the petal width 146 Describing Distributions Unimodal Symmetrical Bimodal Symmetrical 147 Describing Distributions Negative Skew Positive Skew 148 Describing Distributions Kurtosis: Extent of deviation from normal curve in width of curve and thickness of tails Normal Distribution Tall and skinny versus flat and wide 149 Effective and Accurate Use of Graphs Goal of graphs: – To provide an ACCURATE visual representation of your results Not all graphs are created equal! – Some common ways that people use graphs to distort data... 150 Cheating Graphs Goal: To provide an accurate visual representation of your results Problem 1: 45 44 40 42 35 40 30 % Judged 38 % Judged 25 Correctly 36 vs. Correctly 20 15 34 10 32 5 30 0 Lie Truth Lie Truth Type of Story Told Type of Story Told 151 Cheating Graphs Goal: To provide an accurate visual representation of your results Problem 2: 45 45 40 40 35 35 30 30 % Judged 25 Correctly 20 vs. % Judged 25 Correctly 20 15 15 10 10 5 5 0 0 Lie Truth Lie Truth Type of Story Told Type of Story Told 152 Cheating Graphs Goal: To provide an accurate visual representation of your results Problem 3: Failing to adjust for distorting confounds Number Homicides in Canada by Province 1995 200 180 Homicides Per 100,000 in Canada by Province 160 140 120 100 80 60 40 20 0 NL PE NS NB QC ON MB SK AB BC YT NWT 1995 Province Homicides in Canada by Province 1995 14 200 Homicides Per 100,000 12 180 160 140 120 vs. 10 8 Number 100 6 80 4 60 40 2 20 0 0 NL PE NS NB QC ON MB SK AB BC YT NWT NL PE NS NB QC ON MB SK AB BC YT NWT Province Province 153 Graphing Checklist 1. Is there a missing zero? – Vertical axes should include zero or show clearly that it does not with a break in the axis 2. Are both axes labeled? – With both names and units if applicable 3. Is the chart visually confusing or misleading? 4. Are error bars included? 154 Descriptive Statistics: Measures of Central Tendency 155 Descriptive Statistics Numbers that summarize a set of data Two most common ways to summarize data: 1. Measures of central tendency 2. Measures of variability 156 Measures of Central Tendency A measure of the "typical" value in a collection of numbers or a data set. Most often represented by: – Mean – Median – Mode 157 Measure of Central Tendency: Mean Mean (AKA the “average”) Sum of all the scores divided by the total number of scores Population mean:x  Sigma N notation: Σx = add all of Sample mean: x the data entries x (x) in the data n set 158 Example: Finding a Sample Mean Let’s say we take a sample of 7 prisoners and give them the Psychopathy Checklist (PCL). We get the following scores… 38 33 22 36 34 36 26 159 Solution: Finding a Sample Mean 38 33 22 36 34 36 26 The sum of the scores is: To find the mean score, divide the sum of the scores by the number of scores in the sample 160 Measure of Central Tendency: Median The value that lies in the middle of the data when the data set is ordered 1. Rank the data 2. The position of the median is equal to the number of entries plus one divided by two – Odd number of entries? Median is the middle data entry – Even number of entries? Median is the mean of the two middle data entries 161 Example: Finding the Median Looking at our sample of 7 prisoners’ Psychopathy Checklist (PCL) scores… What is the median? 38 33 22 36 34 36 26 162 Solution: Finding the Median 38 33 22 36 34 36 26 First order the data. 22 26 33 34 36 36 38 There are seven entries (an odd number), the median is the middle, or fourth, data entry Or the position is (7 + 1)/2 = position 4 163 Example: Finding the Median 2 What if we have an even number list? What is the median? 38 33 22 36 34 36 26 164 Solution: Finding the Median 2 38 33 22 36 36 26 First order the data. 22 26 33 36 36 38 There are six entries (an even number), the median is the mean of the two middle entries. 165 Measure of Central Tendency: Mode Mode The most frequent value If no entry is repeated the data set has no mode If two entries occur with the same greatest frequency, each entry is a mode – i.e., it is BIMODAL 166 Measure of Central Tendency: Mode Mode The most frequent value If no entry is repeated the data set has no mode If four entries occur with the same greatest frequency, each entry is a mode – i.e., it is MULTI-MODAL 167 Example: Finding the Mode Looking at our sample of 7 prisoners’ Psychopathy Checklist (PCL) scores… What is the mode? 38 33 22 36 34 36 26 168 Solution: Finding the Mode 38 33 22 36 34 36 26 Ordering the data often helps to find the mode. 22 26 33 34 36 36 38 169 Comparing the Mean, Median, and Mode All 3 measures describe a typical entry of a dataset… – Advantage of using the mean: Most common statistic Easily manipulated algebraically The mean is a reliable measure because it takes into account every entry of a data set – Disadvantage of using the mean: Greatly affected by extreme scores (i.e., outliers) Knowledge about individual cases is lost with 170 Comparing the Mean, Median, and Mode All 3 measures describe a typical entry of a dataset… – Advantage of using the median: Little influenced by extreme scores Reasonable estimate of what most people mean by the center of a distribution – Disadvantage of using the median: Slightly less desirable statistical properties than mean May not be good to ignore extreme values 171 Comparing the Mean, Median, and Mode All 3 measures describe a typical entry of a dataset… – Advantage of using the mode: The most frequently obtained score Not influenced by extreme scores – Disadvantage of using the mode: May not represent a large proportion of the scores Ignores extreme values completely 172 Misuse of Central Tendency Statistics Beware of statistics that trade on everyday understanding of “average” to mean “typical” or “common” The arithmetic mean, may or may not be a good indicator of such an understanding Arithmetic mean is greatly influenced by extreme values 173 Mean versus Median Hypothetical “Average” House Prices $398,000 $238,000 $400,000 $392,000 $480,000 $488,000 $500,000 $1,236,000 Average Average $444,500 $588,500 Median: $440,000 174 For normal data, where is… The Mean? The Median? The Mode? 55.87 58.5 61.2 64 66.71 69.42 72.1 8 9 3 Height in inches of women in North America (ages 20- 29) 175 For negatively skewed data, where is… The Mean? The Median? The Mode? Skewed Left Distribution (negatively skewed) The “tail” of the graph elongates more to the left. 176 For positively skewed data, where is… The Mean? The Median? The Mode? Skewed Right Distribution (positively skewed) The “tail” of the graph elongates more to the right. 177 Example: Comparing the Mean, Median, and Mode Find the mean, median, and mode of the sample ages of a class shown below Which measure of central tendency best describes a typical entry of this data set? Ages Are there any in a class outliers? 20 20 20 20 20 20 21 21 21 21 22 22 22 23 23 23 23 24 24 65 178 Solution: Comparing the Mean, Median, and Mode The mean takes every entry into account, but is influenced by the outlier of 65 The median also takes every entry into account, and it is not affected by the outlier In this case the mode exists, but it doesn't appear to represent a typical 179 Appropriate use with scales of measurement Measurement Best indicator Scale of the ‘middle’ Nominal Mode (e.g., gender of participant) Ordinal Median (e.g., class standing based on grades) Interval Mean… if (e.g., temperatures) symmetrical Median… if skewed 180 Ratio Mean… if Important Ethical Principles Belmont Principle 3: Respect for Persons – Autonomy Informed consent – General purpose – Voluntary participation – Indication of risks – Indication of benefits – Freedom to withdrawal – Jargon-free Assent 181 Conducting Ethical Research in Canada Two bodies that provide guidelines for ethical research using human subjects: 1. The Canadian Psychological Association (CPA) 2. Tri-Council Policy Statement 2 (TCPS 2) Most recent (2022) guidelines for ethics standards in research for all who conduct research with humans – CIHR, NSERC, SSHRC – Based on three core principles: 1. Respect for persons : Value for human beings, respect for autonomy-informed consent and protection of vulnerable persons. 2. Concern for welfare :Concern for the physical, mental, spiritual health as well as economic and social circumstances; balancing risk/harm, privacy and responsible handling of human biological materials 3. Justice: treatment of individuals with fairness and equity; marginalized population may need special attention 182 Putting Ethics Into Practice, Part 1 Studying sensitive topics – Sensitive topics aren’t off limits Think creatively/smaller Use sound research design – Example Cheating/stealing Discussion Question: What are some ethical ways to study this topic? 183 Open Science Practices Scientific integrity Open science Science must be self-correcting. 184 Case Study: Stapel “I failed as a scientist. I adapted research data and fabricated research. Not once, but several times, not for a short period, but over a longer period of time.” ~ Diederik Stapel 185 Replication and Reproducibility Direct replication Conceptual replication Reproducibility The reproducibility project only replicated 40% of effects. Reasons studies fail to replicate: – Social changes – Different participants – Accidental errors 186 Questionable Research Practices Questionable research practices (QRP) Examples of QRPs: A researcher collects data and tests several hypotheses but only reports the ones that show significant results, ignoring others that didn’t work out. QRPs are common 187 Three Badges of Open Science Preregistered Open data Open materials 188 Scientific Integrity The Ethical Presentation of Findings – Plagiarism: Representing others’ ideas as your own, or without giving proper credit Examples (https://www.indiana.edu/~istd/examples.html) Self-plagiarism – Paraphrasing: Summarizing others’ ideas in your own words while providing a proper citation Discussion Question: Is it ethical to take work from one class and use it in another? Why or why not? 189 Activity: Powers of Paraphrasing Avoiding Plagiarism Through Paraphrasing – Original Thompson, A. E., & O’Sullivan, L. F. (2012). Gender differences in associations of sexual and romantic stimuli: Do young men really prefer sex over romance? Archives of Sexual Behavior, 41(4), 949– 957. doi:10.1007/s10508-011-9794-5. We expected that men would show a stronger preference for sexual over romantic stimuli, in line with theory and research regarding gender differences. Contrary to our hypothesis, however, this study revealed that both men and women showed a stronger automatic preference for romantic stimuli as compared to sexual stimuli. There was, however, a gender difference in the degree to which the romantic stimuli were preferred. Specifically, women had a stronger automatic preference for romantic stimuli than did men, which corresponds to both evolutionary and sociocultural theories regarding sexual roles. – How could you paraphrase this? 190 Using Animals in Research 191 Using Animals in Research Considerations: 1. Does animal research translate to humans? 2. How ethical is it to use animals in research? – Can’t give informed consent 192 Using Animals in Brain-Behaviour Research Benefits of animal research (humans): – Polio, smallpox, diptheria, cholera, measles are no longer a serious threat. – Early diagnosis and treatment of disorders. – Surgical and treatment methodologies. 193 Using Animals in Brain-Behaviour Research CCAC is dedicated to: – Enhanced animal care – Education – Voluntary compliance – Code of ethics The three Rs 1. Replacement 2. Reduction 3. Refinement 194 Using Animals in Brain-Behaviour Research Four basic guidelines of the CCAC: 1) The use of animals in research, teaching, and testing is acceptable only if it contributes to the understanding of environmental principles or issues, fundamental biological principles, or development of knowledge that can reasonably be expected to benefit humans, animals, or the environment. 195 Using Animals in Brain-Behaviour Research Four basic guidelines of the CCAC: 2) Optimal standards for animal health can care result in enhance credibility and reproducibility of experimental results. 3) Acceptance of animal use in science critically depends on maintaining public confidence in the mechanisms and processes used to ensure necessary, human, and justified animal use. 196 Using Animals in Brain-Behaviour Research Four basic guidelines of the CCAC: 4) Animals are used only if the researcher’s best efforts to find an alternative have failed. Researchers who use animals employ the most humane methods on the smallest number of appropriate animals required to obtain valid information. 197 Writing an APA formatted Paper and APA 7th Edition 198 Parts of a Lab Report/Research Paper Title Page Abstract Introduction Method Results Discussion References Appendices (if any) 199 APA Title Page 7th ed. No running head included on any page, including title page 200 Introduction 1.Introduce your research topic – Explain/define key concepts and ideas – Briefly identify your research question/purpose of your study – Review the most important and relevant background research 2. Justify your research question 1. Why is your research question important? 1. e.g., advancement of theory, issue of public concern, unresolved issue stemming from previous research 3.Set up current study Briefly describe focus of current study 201 – State and provide empirical support for your hypotheses Method Participants – Sample size, gender breakdown, mean age and standard deviation, sample characteristics (e.g., university students), method of recruitment, reimbursement, drop- out rates Materials – Describe questionnaires, tests, other materials used (e.g., videos, photos, etc.) Procedure – Describe sequence of events in which participants took part, including any randomization methods and instructions Provide enough detail to ensure your study can be replicated 202 Results Should include both descriptive statistics (typically means and SDs) and inferential statistics (t-test, ANOVA, etc.) Make sure to identify your levels of your IV, your DV, AND the direction of your effect rather than simply saying that the conditions were significantly different – e.g., Students who used active study strategies (M = 95.92%, SD = 3.45%) scored significantly higher test scores than those who used passive study strategies (M = 65.32%, SD = 4.45%), t(30) = 8.94, p =.002. 203 - No border Axis just long enough to - No gridlines accommodate bar length - White background Figure 1 Changes in Test Scores as a Function of Study 100 Strategy 80 Mean Test Score Don’t 60 forget Y-axis: DV your 40 error 20 bars! If possible, do not 0 truncate your Active Passive graph Type of Study Strategy Employed X-axis: IV 204 Discussion Restate the purpose of your study Provide a short summary of your findings Situate your findings within past research ‾ How do your findings fit with past research? ‾ If your findings are inconsistent, provide potential reasons for this inconsistency What are the implications of your findings? 205 How to Cite Sources in the Body of Your Paper Known as “In Text Citation” Authors’ last names followed by the year – If you cite your authors within a sentence: Byron and Hunt (2017) found that blue marbles are brighter than green marbles. – If you cite your authors at the end of the sentence: Green marbles are brighter than blue marbles (Byron & Hunt, 2017). 206 APA Citation 7th Edition: General Rules  Double-space your paper, including the reference list, with 1” margins  Use a font consistently throughout. Recommendations include sans serif fonts such as 11-point Calibri, 11-point Arial, or 10- point Lucida Sans Unicode; and serif fonts such as 12-point Times New Roman, 11-point Georgia, or 10-point Computer Modern  Include a page header at the top of every page with the page number, flush right  Format reference list entries with a hanging indent  Arrange reference list entries in alphabetical order by the surname of the first author or by title if there is no author. 207 In Text Citations For 7th edition for 3 or more authors: Every time (including the first time) you refer to the same study, you write “et al.” In text: Maeder et al. (2016) In parentheses: (Maeder et al., 2016) 208 In Text Citation One work by one author  Author’s name cited in-text: Petrusic (1992) reported...  Author’s name cited in parentheses: In a study on the comparison process (Petrusic, 1992)… One work by two authors  Rasmusson and Friedman (2002) found...  A study on gender issues in PTSD (Rasmusson & Friedman, 2002) showed... One work by three or more authors  Subsequent citations: ‒ Petrusic et al. (1998) found that... ‒ A study of the psychophysics of visual memory (Petrusic et al., 1998)... 209 In Text Citations Two or more works by the same author named in the same reference  Past research (Petrusic, 1984, 1992) has shown... Two or more works by different authors named in the same reference  Past research (Heschl, 2001; Noonan & Johnson, 2002; Wolchik et al., 2000) has shown... Citing a secondary source (Note: use sparingly; primary sources are strongly preferred)  In Smith’s 1998 study (as cited in Rasmusson & Friedman, 2002)...  Experimental research (Smith, 1998, as cited in Rasmusson & Friedman, 2002) has shown... ‒ Do not include Smith (1998) in the reference list. ‒ Do include Rasmusson & Friedman (2002). 210 In Text Citations Direct quotations (Note: use sparingly; paraphrasing is strongly preferred) When you cite direct quotations, include the page number of the quote.  Stereotypes have been defined as “generalized and usually value-laden impressions that one social group uses in characterizing members of another group” (Lindgren, 2001, p. 1617). Electronic sources without page numbers  (Myers, 2000)  If there is no author, use the title in italics (or a short form of the title, if it is lengthy) and the year: (What is alcohol poisoning?, 2000)  If there is no date, use n.d. instead: (Myers, n.d.) 211 APA Citation 7th Edition: General Rules  Use only the initial(s) of the author’s given name, not the full name.  If the reference list includes two or more entries by the same author(s), list them in chronological order (oldest first).  Capitalize only the first letter of the first word in the article title and subtitle.  Italicize journal titles and volume numbers. Do not italicize issue numbers.  References cited in text must appear in the reference list and vice versa. The only exceptions to this rule are personal communications and secondary sources, which are cited in text only and not included in the reference list. 212 References Remember Primary sources (i.e., journal articles describing an original study) almost always superior to secondary sources (i.e., book chapters) ALL sources cited in paper must be included in References section Authors listed in alphabetical order by first author’s last name 213 References Agarwal, S. M., Shivakumar, V., Kalmady, S. V., Danivas, V., Amaresha, A. C., Bose, A., Narayanaswamy, J. C., Amorim, M., & Venkatasubramanian, G. (2017). Neural correlates of a perspective-taking task using in a realistic three-dimensional environment based task: A pilot functional magnetic resonance imaging study. Clinical Psychopharmacology and Neuroscience, 15(3), 276–280. https://doi.org/10.9758/cpn.2017.15.3.276 British Columbia Ministry of Health. (1999). Provincial Health Officer’s Annual Report. Canada (Attorney General) v. Canadian Pacific Ltd. (2001), 217 D.L.R. (4th) 83 (B.C.C.A.). Heschl, A. (2001). The intelligent genome: On the origin of the human mind by mutation and selection. Springer-Verlag. 214 Some General Tips for Report Writing AVOID use of direct quotes! – Put into your own words (i.e., paraphrase) whenever possible! Make sure you’ve actually read ALL articles you cite! Unsure of how to write up something? – Look at other journal articles! Title, Abstract, and References go on separate pages but no page separations between other sections of paper 215 Sampling and Research Designs 216 Sampling Sampling from a population of interest 217 Probability Sampling Simple random: Each member of the population has equal chance of being selected Systemic: Based on intervals, i.e every nth person Cluster: population ids divided into groups or clusters that are then randomly Stratified random: Population is divided into groups (strata) and random samples are selected from each stratum 218 Non-Probability Sampling Convenience: sampling based on ease of access Snowball: sample group grows as participants recruit further participants Quota: Divide the population into strata and use judgment to choose participants based on a specific proportion or quota 219 Nonexperimental (or Correlational) Designs Nonexperimental design Explanatory (or predictor) variable Criterion (or response) variable: The outcome variable in nonexperimental designs. Why use a correlational design? 220 Ethics 221 What Are Ethics? Ethics involves the application of moral principles concerning what an individual considers right or wrong to help guide decisions and behaviours. Ethical Perspectives – Utilitarian perspective – Altruistic perspective – Egoism 222 Important Ethical Principles, Part 1 The Nuremburg Code (1947) The Belmont Report (1978) – Belmont Principle 1: Beneficence and Nonmaleficence Cost-benefit analysis Beneficence Nonmaleficence – Confidentiality – Anonymity Types of harm – Physical harm – Psychological harm Cost of not doing the research 223 Important Ethical Principles, Part 2 Belmont Principle 2: Justice – Justice Examples: – Tuskegee study Clinical equipoise – An Ethical Dilemma 224 Last Class Choosing a good research question – Based on interest (the answer is in doubt, the answer fills a gap in the research literature, and the answer has important practical implications) and feasibility (time, money, resources, etc…) Searching the literature and the peer review process Formulating a hypothesis 225 Strategies for Generating Hypotheses Introspection Find the exception to the rule A matter of degree Change the directionality 226 Evaluating Your Hypothesis Does it correspond with reality? – It should be consistent with past research. Is it parsimonious? – Occam’s razor How specific is it? – The Barnum effect Is it falsifiable or refutable? 227 Important Decisions to Make What are the key concepts in the hypothesis? – Variables – Constant How will I define these concepts? – Conceptual definition – Operational definition – There are many ways to operationalize the same conceptual variable. Which research decision should I use? – “Why” versus “what” questions. 228 Experimental Designs Experimental design Independent variables (IVs) – Levels Dependent variables (DVs) 229 IDENTIFYING INDEPENDENT AND DEPENDENT VARIABLES 230 Variables Can be categorical or quantitative Need to be operationally defined 231 Term Report 232 Always Begin with a Good Research Question Research questions are often inspired by everyday experiences and observations. – Choose an “interesting research question” – Choose a “feasible research question” – Research questions often begin as more general research ideas focusing on some behavior – Arise from informal observations, practical problems, previous research Empirical versus nonempirical questions Direct versus indirect observations 233 Why Search the Literature? Part 1 Discovering what others have learned will help us to ask a better, more specific question that will continue to advance our knowledge. 234 Why Search the Literature? Part 2 Peer review – The review process is usually “blind.” Types of articles: – Research report (or research article) – Systematic review – Meta-analysis Downsides? 235 Searching the Literature You can find peer-reviewed articles using databases, such as PsycINFO and PsycARTICLES. 236 Constructing a Hypothesis Scientific Law Scientific Theory – Psychology relies much more on theories than laws. Hypothesis 237 Last Class Why research methods are important – Heuristics (mental shortcuts) and fallacies – Problems with introspection Pleasure paradox 238 Introduction to Research Methods Why research methods Characteristics of a good are important scientist Benefits of learning the scientific method 239 Characteristics of a Good Scientist Skepticism Open-mindedness Objectivity Empiricism – Empirical vs. nonempirical research Creativity Communication – Replication 240 Why Learn about the Scientific Method? Value and produce both basic and applied research. – Basic research: Expanding knowledge on a topic to build what we already know by developing theories to explain phenomena in our world – Applied research: Has a goal of solving a practical problem or examining the real-world implications of a [particular theory 241 Become a Better Consumer of Research  Learn to deal with science denialism  Identify when science deniers use FLICC to counteract denialism: 1. Fake experts or presenting information from questionable sources (people or institutions) 2. Logical fallacies or arguments that use errors in reasoning 3. Impossible expectations or creating unrealistic goals of certainty before believing a fact 4. Cherry picking or selecting only the data that supports a claim 5. Conspiracy theories or conjuring a secret schene to explain straightforward findings  Spot pseudoscience 242 The Scientific Method: Developing a good research idea and question The Research Process Psychology 2013 Quantitative Research Methods Fall 2024 245 Last Class Introduction to the course and course outcomes 246 ? QUESTION How much do you rely on objective evidence when judging others, as opposed to relying on your gut instincts? A. I rely mostly on objective evidence. B. I rely about equally on objective evidence and gut instincts. C. I rely mostly on gut instincts. Introduction to Research Methods Why research methods Characteristics of a good are important scientist Benefits of learning the scientific method 248 Something to Think About What is the best way to figure out what a person is really like? 249 Suppose You Had to Choose a New Roommate from a Group of Strangers... What would be the best strategy for selecting someone? 250 Would Analyzing Someone’s Handwriting Give You Clues about Their True Personality? 251 Thinking Differently Research methods and statistics courses focus more on “know- how” than “know- what.” Researchers do not rely on their gut instincts. – They use the scientific method to find out the truth. 252 Heuristics Are Useful... Most of the Time Availability heuristic: A mental shortcut strategy for judging the likelihood of an event or situation to occur based on how easily we can think of similar or relevant instances. Representativeness heuristic Using heuristics is not necessarily a bad thing, because they help us avoid becoming overwhelmed. - the problem is most people are not aware of their limitations. 253 Natural Flaws in Thinking Better-than-average effect: The tendency to overestimate your skills, abilities, and performance in comparison to those of others Overconfidence phenomenon: Overestimating one’s knowledge, accuracy, or abilities regardless of how they compare to others. 254 Natural Flaws in Thinking, Part 2 Hindsight bias Confirmation bias Focusing effect 255 Looking within Self to Gain Knowledge Introspection “What You See Is All There Is” Phenomenon Pleasure Paradox 256 Problem of Belief Perseverance 257 Anecdotal Versus Scientific Evidence We tend to overvalue personal experiences and anecdotes when drawing conclusions. Law of Small Numbers Problem with Outliers – Keith Stanovich’s “person-who” statistics Scientific Method 258

Psychology 2103 Quantitative Research Methods Fall 2024 PDF

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue