Unit 2: Research methods and statistics-complete

1. Operational definition
def: a precise concrete description of how every variable in a psychological study will be measured, manipulated, or changed so that the study can be replicated (these are important because many of the concepts psychologists study are hard to measure and can be measured in more than one way)
Ex: In measuring how premarital education benefited martial quality, “martial quality” was operationally defined as a person’s 1-5 rating in response to several questions about their satisfaction with their marriage as part of a survey

2. *Nautalistic observation
def: a research method that involves systematically recording and observing the behavioral patterns of animals or humans on display in natural settings
Ex: To collect data about the pace of life in different countries (the amount of time it takes a pedestrian to walk 60 ft in a downtown city street), researchers timed how long it took 35 male and 35 female pedestrians to walk this far in each city without them noticing. Based on their observations, they found that the pace of life was fastest in Ireland and slowest in Brazil.
Ex: if you recorded teenagers talking at a restaurant you could study what teenagers talk about
–1. allows psychologists to study behaviors that would be logistically or ethically impossible to create in a lab such as bullying behavior in children (could observe drug use in teenagers in bars)
–2. theories developed from this research method can be generalized more easily than theories developed from other research methods since observations are based on behaviors observed in natural settings as opposed to behavoir observed in manipulated or staged situations
–1. the experimenter cannot control all the extraneous variables that could be influencing the behavior they are observing meaning that they cannot effectively prove a cause and effect relationship between two variables
–2. Different observers may draw different conclusions after witnessing the same behavior in a natural setting
–3. If people discover what you are doing, they might try to shock you.
–4. Could take forever to observe bahavior that is relevant

3. Case study
def: an intensive in-depth investigation of an individual or small group of individuals such as a family used in rare or extreme circumstances which involves gathering a great deal of information from numerous sources to construct an accurate picture of the person or people being studied
Ex: by extensively studying the brain and visual abilities of Mike who regained his sight after decades of blindness, psychologists learned useful information about how brain and visual development works in normally sighted individuals
–1. By studying these rare and extreme phenomena, psychologists can gain additional insights into how normal behaviors work
–2. Performing cases studies on psychotherapy clients, allows clinical psychologists to treat them more effectively
–1. Hard to identify which variables caused or influenced specific behaviors
–2. on account of the the rare and extreme nature of the behaviors studied in case studies, their results often do not apply to the general population rendering them a waste of time and resources.(can’t get good data)
–3. Observer bias because you get too close

4. *Survey
def: a questionnaire or interview designed to investigate the opinions, behaviors, or characteristics of a particular group by asking them a structured set of questions about their experiences, beliefs, behaviors and attitudes (can be conducted with pencil and paper, virtually, as well as in person or on the telephone)
Ex: Political polls are surveys conducted by phone which ask people in specific political district a structured set of questions related to their beliefs and attitudes about political candidates or issues
–1. because surveys can be administered remotely using them as research tools allows researchers to gather data about a larger number of people than possible in when using any other research methods
–2. because they are standardized, the results gained from surveys are easier to compile that the results from other research methods
–1. people may answer the questions dishonestly
–2. data may be incorrect since the way a subject interprets a question often differs from what the researcher intended the question to convey

5. Representative sample
def: a sample of people that is representative of (closely aligns or matches) the population being studied in terms of relevant characteristics such as age sex, race, martial statues and education level (the best results are found when survey respondents make up a representative sample)
Ex: the NHSLS was distributed to a representative sample of adults in the US ages 15-59, because the distripution of relevant characteristics such as race, gender, martial statues, and education level among the people who responsed to the survey closely aligned with distripution of these relevant characteristics among american adults as a whole in this age range

6. Random selection
def: process in which subjects of research are selected randomly from a large group where every group member as an equal chance of being selected (often the demographics of people in a survey are decided beforehand but the exact individual who make up this demographic are chosen through random selection)
Ex: A psychologists studying mental illness among hispanic women gathers the names of 50,000 Hispanic women and places them in a hat. In order to select 200 subjects to participate in her study through random selection, she picks names randomly out of the hat until 200 names have been picked.

7. Correlation coefficient
def: numerical indicator of the strength and the direction of the relationship between two varaibles
which always falls between ±1.00 (the closer a correlation coefficient is to |1.00|, the stronger the correlation/association between those two factors)
Ex: If researchers found that hours of sleep per night and high school GPA had a correlation coefficient of +9.00, that would indicate that these variables were strongly correlated with one another meaning that in almost every case, as the number of hours a students sleeps increases their GPA does as well

8. Positive/negative correlation
positive correlation (directly propertional): a finding that two facters vary systematically the same direction either increasing or decreasing together
–Ex: because Karen Dill and Craig Anderson found that as the amount of time an individual spent playing violent video games increased their aggression scores on personality tests also increased, they concluded that these two factors were positively correlated
negative correlation (indirectly/inversely proportional): a finding that two factor vary systematically in opposite directions meaning that as one factor increases the other always decreases
–Ex: because Karen Dill and Craig Anderson found that as the amount of time an individual spent playing violent video games increased their college GPA decreased, they concluded that college GPA and the amount of time spent playing violent video games were negatively correlated

9. *Experimental method
def: a research methods used to demonstrate a cause and effect relationship between two variables (shows that a changes in one variable causes changes in the second) by purposely manipulate one factor (IV) thought to produce a change in another factor (DV)
–always involves an experimental group who are exposed to all of the experimental conditions including the IV and a control group who are exposed to all of these same condtions excepted for the IV. The results from the control group serve as the baseline with which to compare the experimental group.
Ex: If a researcher wanted to measure how exposure to loud noises affected student test scores, they would divide their test subjects into two group and put them in seperate rooms to take tests. Next, they would manipulate the IV by playing drums in one room and having nothing playing in the other. To measure how much the IV noises affected the DV (subject’s test scores), they would compare the test scores of the subjects in the two rooms.
–1. with this research method, because researchers can control most of the confounding variables that could interfere with an experiment (must be careful with all of these), using it allows them to effectively prove cause and effect relationships (that change in the IV are directly responsibly for the changes observed in the DV)
1. Only experimental method which can prove cause and effect relationships
–2. variety of variations available: researchers can tailor the details of the experiment so that it addresses what they want to study specifically
–1. because these experiments take place in labs, the observed results are less applicable to real world situations than other methods since people often act differently when in a lab setting than in real life
–2. because these experiments are often time consuming and expensive to conduct, this researchers method only allows research to study a limited amount of people (can’t take charge of all controls)
–3. Very subjective to demand characteristics

10. Independent variable/Treatment Vairable (IV)
def: the purposely manipulated factor in an experiment conducted with experimental methods thought to produce change in the other factor (DV)
Ex: In an experiment focused on studying how the color of an office influences worker productivity where one group performs a task in a yellow room while another performs the same task in a blue room, the color of the room would be considered the independent variable

11. Dependant variable (DV)/Outcome Variable
Def: the factor observed and measured for change in in an experiment conducted with experimental methods which is thought to be influenced by the other factor (IV)
Ex: In an experiment where researchers are studying whether first-born or second-born children learn to speak at a younger age by asking parents to fill out a survey about their children developments, the dependent variable would be the age the children started speaking.

12. Confounding/ Extraneous variables
def: variables other that the one being studied in an experiment that if not controlled can affect its outcome by influencing changes in the DV
Ex: In Milligram’s experiment, telling subjects before they engaged in the experiment that the teacher never actually received any shocks would have been a confounding vairable since it would have greatly affected the actions of teachers in the experiment and the results of the experiment as a whole

13. Placebo effect
Def: physcial changes attributed to a person’s incorrect belief in the effectiveness of a certain drug or procedures rather than the actual treatment or procedure
Ex: In Equus, taking a baby aspirin causes Alan to tell the truth about his troubled childhood because he believes that the drug forces people who take it to tell the truth

14. Random assignment
Def: method assignment in an experiments where because the same method is used to assign participants to the different groups, each participant has an equal change of being assigned to either the experimental or control group (helps to ensure that any potential differnces among the participant are spread out evenly among the groups as well as minimizes possibilities of bias in group assignment)
Ex: If you were curious about the effects of eating an apple a day on blood pressure, you might design a study where one group (the experiental group) eat an apple every day while the other group (the control group) does not. To ensure that your results are accurate by ensuring that any potential differences such as prior blood pressure conditions are spread out evenly among both of the groups, you should randomly assign each of the subjects to one of the two groups

15. Double-blind technique
def: experimental technique where neither the participants nor the researcher know who is part of the control and experimental groups in order to prevent researchers from accidentally becoming confounding variables in the study
Ex: in the ginko experiment where the control group were given fake pills and the experimental was given pills containing ginko, the double blind technique was utilized since the researchers who interacted with the participants did not know which participants were given the real or fake ginko pills (the researcher who knew the group assignments did not interact with or evaluate the participants)

16. Demand characteristics
Def: subtle hints or cues that researchers accidentally display when dealing with some participants and not others that can bias the outcome of the study by communicating the behavior or response that is expected of the participants
Ex: In an experiment, overseen by Margaret Jean Intons-Peterson but run by her assistants where subjects had to complete 1 math and 1 logic problem based on the instructions given by the assistants, Intons-Peterson told one of her assistants that the subjects should do better on the math problem and the other that the subjects should do better on the logic problem. Because each assistance unknowingly read out the directions to the problem they thought the subject would do better on more slowly and clearly, the subject did better on that problem in most cases. This experiment illustrates that thesubtle cues a researcher demonstrates within an experiment greatly affect its outcome.

17. Skewed distribution (+and-)
17. Skewed distribution (+and-)
def: a distribution whose graph is assymetrical (if you drew a line through th midline of the x-axis, more scores would fall on one side of the distribution than the other)
positive skewed distribution: distribution where on its graph more scores fall on the lower end (to the left of the line) than at the high end meaning that most scores in the distribution are low scores
Negative skewed distribution: distribution where on its graph more scores fall on are high end of distribution (to the right of the line) than the low end meaning that most scores in the distribution are high scores
Ex: The above below represent about how many hours of exercise people attain per week gathered in a study about weight gain. Because most of the scores are on the lower end of the graph to the left of the midpoint of the x-axis, this must be a positively skewed distribution demonstrating the most people do not exercise very much on a regular basis.

18. Mode
def: the most frequently occurring score in a distribution representing graphically as the value of the x-axis with the greatest frequency
Ex: in the distribution {1,1,1,10, 20, 30}, 1 is the mode because it appears 3 times while the other numbers only appear once

19. Median
def: the score that falls in the middle of a distribution (same distance from lowest and highest number in the sequences) when it is written out in numerical order so that the same number of scores lie on each side of it (If there is an even number of numbers in the distribution, the median is the average of the two middle numbers)
Ex: In the 29 number sequence following
{0,0,0,0,1,1,1,1,1,1,1,2,2,2,2,2,2,2,3,3,3,3,4,4,5,5,6,7}, 2 is the median because there are 15th terms between it and either side

20. Mean/arithmetic average (X ̅)
def: the sum of the set of scores in a distribution divided by the total number of scores in the distribution which serves as the most representative measure of its central tendency (“typical” score)
–Very susceptible to extreme scores because their presence in a distribution pulls the mean in one direction
formula: Formula: X ̅=(∑X)/N where ∑ represents the Greek letter which means “sum of”, X represents the scores in the distribution, and N represents the total amount of scores in the distribution
Ex: Finding the mean of the 6 score distribution {5,4,12,2,1,6} requires adding all the scores together attaining a sum value of 30 (5+4+12+2+1+6= 30) and then dividing that number by the amount of scores in the distribution (6) attaining a mean of 6 ( X ̅=30/5=6)

21. Range (R)
def: a measure of vitality (how much scores differ from one another) found by subtracting the lowest scores in a distribution from the highest (gives a limited amount of information because it depends on the two most extreme scores in the distribution)
formula: A – I =R where A represents the maximum value and I the minimum value in the distribution
Ex: If after 6 months in a weight loss program, the weight of the lightest participant is 95 pounds and the weight of the heaviest participant is 115 pounds, the range of weights of the participants after the program is given by the formula 115-95 = 20 lbs

22. Standard deviation (SD)
22. Standard deviation (SD)
def: a measure of vitality (how much scores differ from one another) of distributions expressed as the square root of the sum of all squared deviations (distances from each score to the mean) from the mean of a distribution divided by the number of scores in the distribution
Formula: given above where ∑ represents a Greek letter which means “sum of”, X represents the scores in the distribution, M represents the mean of the distribution, and N represents the total number of scores in the distribution
Ex: Finding the standard deviation of the following 15 term distribution {155, 149, 142, 138, 134, 131, 127, 125, 120, 115, 112, 110, 105, 102, 95} requiring finding the mean of the distribution (124), subtracting the mean from all of the scores to find the derivations, squaring each of the derivations, and adding them all together to attain a sum of 4,388. At this point, the standard of derivation of the distribution can found by solving the equation SD = √(4,388/15) = 17.10

23. Z-score (z)
def: a number expressed in standard deviation units that gives the deviation of a specific score in a distribution from the mean
formula: z=(X-X ̅)/σ where X represents a score in the distribution, X ̅ represents the distribution’s mean, N represents the total number of scores in the distribution, and σ represents the standard deviation of the distribution
Ex: after finding the standard deviation (17.10) and mean (124 lbs) of the distribution, the z-score of a weight of 149 lbs within a distribution which represent the weights of 15 participants after undergoing a 6 month weight loss program can be attained by solving the equation z=(149-124)/17.10=1.46

24. Normal distribution/Normal Curve
24. Normal distribution/Normal Curve
def: a bell shaped curve which represents the distribution of individual scores in which most scores cluster around the average score (as scores become more extreme fewer instances occur)
Ex: Because most people score between 85-155 on IQ tests, while less than 1% of people score higher than 145 or lower than 55, a distribution representing the frequency of IQ scores can be classified as a normal distribution which peaks between the scores of 85 and 155.

25. Scatter plot/
scatter diagram
25. Scatter plot/
scatter diagram
def: graph that represents the relationship between two variables
–Scatter plots of positive correlations have positive slopes while scatter plots of negative correlations have negative slopes
–the closer the points are to a diagonal line moving either upward of downward, the strong the correlation between the two variables
Ex: Because the lines in the scatter plot above resembles a diagonal line with a positive slope, the correlation the graph represents must be positive. Additionally, since many of the points do fit into the shape of a positively sloped diagonal line, the correlation the graph represent must be strong.