psyc research methods exam 3

coveriance

= correaltion

selection-history

A threat to internal validity in which a historical or seasonal event systematically affects only the participants in the treatment group or only those in the comparison group, not both.

AUTOCORRELATIONS

Autocorrelations: the correlation of each variable with itself across time. - The next step was to evaluate the correlation of each variable with itself across time = they determine the correlation of one variable with itself, measured on two different occasions ex= For example, the Brummelman team asked whether mothers' overvaluation at Time 1 was associated with mothers' overvaluation at Times 2, 3, and 4 - idicative of re-test test valibility So far, so good. However, cross-sectional correlations and autocorrelations are generally not the primary interest.

CHAPTER 8:

BIVARIATE CORRELATIONAL RESEARCH

the 3 branches of life

Bacteria, Archaea, Eukarya

BIVARIATE CORRELATIONS

Bivariate correlations: associations that involve exactly two variables Associations between quantitative variables Associations between quantitative and qualitative variables \\ An association claim describes the relationship found between two measured variables. A bivariate correlation, or bivariate association, is an association that involves exactly two variables.

CHAPTER 11:

CONFOUNDING AND OBSCURING VARIABLE

Practice effects

Order effects can include practice effects, also known as fatigue effects, in which a long sequence might lead participants to get better at the task or to get tired or bored toward the end.

Telomeres

Repeated DNA sequences at the ends of eukaryotic chromosomes.

STATISTICAL VALIDITY QUESTION 3: HAS IT BEEN REPLICATED?

The effect size and 95% CI provide important information about how strong the relationship might be. Another step in estimating the population association is to conduct the study again (a process called replication) and find multiple estimates.

SELECTION EFFECTS—RANDOM ASSIGNMENT

Well-designed experiments often use random assignment to avoid selection effects = random assignment The use of a random method (e.g., flipping a coin) to assign participants into different experimental groups. - Assigning participants at random to different levels of the independent variable—by flipping a coin, rolling a die, or using a random number generator—controls for all sorts of potential selection effects

WITHIN-GROUPS DESIGNS

repeated-measures design. - Concurrent-Measures Design

epigentics

the study of environmental influences on gene expression that occur without a DNA change

temporal precedence

one of three criteria for establishing a causal claim, stating that the proposed causal variable comes first in time, before the proposed outcome variable

within-groups design

or within-subjects design, each person is presented with all levels of the independent variable. - within- withney experced the whole rainbow - within-groups design if they had asked each participant to take notes on two videos—using a laptop for one and handwriting their notes for the other.

ordinal scale

orbinal - orbit - eye - ranked from the body - head torso , legs , not equal lents A quantitative measurement scale whose levels represent a ranked order, and in which distances between levels are not equal (e.g., order of finishers in a race) Objects or individuals are categorized, and the categories (still qualitative data) form a rank order along a continuum - Sometimes referred to as ranked data Properties of identity and magnitude Class rank, Letter grade Mathematical Operations on Ordinal Data:Rank order

THREE TYPES OF QUANTITATIVE VARIABLES

ordinal scale - interval scale - ratio scale

Multiple-Regression Results from a Study Predicting Academic Success from Frequency of Family Meals and Parental Involvement

*p < .05, meaning the result is statistically significant and the 95% CI does not include zero.

THE REALLY BAD EXPERIMENT: ONEGROUP PRETEST/POSTTEST DESIGN

-A general diagram of the really bad experiment, or the one-group, pretest/posttest design. Unlike the pretest/posttest design, it has only one group: no comparison condition.

COMBINED THREATS

. However, three more threats to internal validity—observer bias, demand characteristics, and placebo effects —might apply even for designs with a clear comparison group.

RULING OUT THIRD VARIABLES WITH MULTIPLE-REGRESSION ANALYSES

/internal valitbity - Multiple regression (also known as multivariate regression): helps address questions of internal validity by ruling out some third variables =A statistical technique that computes the relationship between a predictor variable and a criterion variable, controlling for other predictor variables. Also called multivariate regression. - which can help rule out some third variables, thereby addressing some internal validity concerns. - If the researchers had stopped there and measured only these two variables, they would have conducted a bivariate correlational study. . However, they also measured several other variables, including the total amount of time teenage participants spent watching any kind of TV, their age, their academic grades, and whether they lived with both parents. By measuring all these variables instead of just two (with the goal of testing the interrelationships among them all), they conducted a multivariate correlational study. - - Groundbreaking research suggests that pregnancy rates are much higher among teens who watch a lot of TV with sexual dialogue and behavior than among those who have tamer viewing tastes

SIX POTENTIAL INTERNAL VALIDITY THREATS IN ONE-GROUP, PRETEST/POSTTEST DESIGNS

1. Maturation threats to internal validity 2. History threats to internal validity 3. Regression threats to internal validity 4. Attrition threats to internal validity 5. Testing threats to internal validity 6. Instrumentation threats to internal validity

ADVANTAGES OF WITHIN-GROUPS DESIGNS

1. Participants in your groups are (1.)equivalent because they are the same participants and (2.)serve as their own controls. -(1)The only difference between the two conditions can be attributed to the independent variable -(2) The idea of "treating each participant as his or her own control" also means matched-groups designs can be treated as within-groups designs. As discussed earlier, in a matched-groups design, researchers carefully match sets of participants on some key control variable (such as GPA) and assign each member of a set to a different group 2. Within-groups designs require fewer participants than other designs.

DISADVANTAGES OF WITHIN-GROUPS DESIGNS

1. Potential for order effects — Solution: counterbalancing 2. Might not be practical or possible 3. Experiencing all levels of the independent variable (IV) changes the way participants act (demand characteristics)= =A cue that can lead participants to guess an experiment's hypothesis is known as a demand characteristic, or an experimental demand. Demand characteristics create an alternative explanation for a study's results.

Systematic variability

==In an experiment, a description of when the levels of a variable coincide in some predictable way with experimental group membership, creating a potential confound. See also Systematic Variability Is the Problem. -You need to be careful before accusing a study of having a design confound. -Not every potentially problematic variable is a confound = . Did the cheerful models work only in the effort condition and the reserved ones only in the no-effort condition? = Then it would be a design confound = uniformatity in a varibble that is not the IV unsystematic variability In an experiment, a description of when the levels of a variable fluctuate independently of experimental group membership, contributing to variability within groups. If those in the longhand group all happened to be very interested in the lectures and those in the laptop group were all uninterested, level of interest would vary systematically with the notetaking condition and would be a confound. - But if some participants in each condition were interested and some were not, that would be unsystematic variability and would not be a confound.

Latin square

A formal system of partial counterbalancing that ensures that each condition in a within-groups design appears in each position at least once. - A counterbalancing technique to control for order effects without using all possible orders. A more formal way to use incomplete counterbalancing that limits the number of orders of conditions.

PLACEBO EFFECTS

A placebo effect occurs when people receive a treatment and really improve—but only because the recipients believe they are receiving a valid treatment - can be a pill or in psy a fake therpay talk

interval scale

A quantitative measurement scale that has no "true zero," and in which the numerals represent equal intervals (distances) between levels (e.g., temperature in degrees)= pH, SAT score (200-800), credit score - Intervals between the numbers (quantitative data) on the scale are all equal in size!!!!! Distance between numbers is meaningful Scale permits mathematical operations of addition and subtraction. Properties of identity, magnitude, and equal unit size

one-group, pretest/posttest

A researcher recruits one group of participants; measures them on a pretest; exposes them to a treatment, intervention, or change; and then measures them on a posttest. This design differs from the true pretest/posttest design you learned in Chapter 10 because it has only one group, not two. There is no comparison group. Therefore, a better name for this design might be "the really bad experiment."

Ratio scale

A scale (quantitative data) in which, in addition to order and equal units of measurement, an absolute zero indicates an absence of the variable being measured Scale permits mathematical operations of addition, subtraction, multiplication, and division (i.e. you can say something is twice as big as something else). a quantitative scale of measurement in which the numerals have equal intervals and the value of zero truly means "nothing" A quantitative measurement scale in which the numerals have equal intervals and the value of zero truly means "none" of the variable being measured - Properties of identity, magnitude, equal unit size, and absolute zero Age Weight Height Sales Figures Ruler measurements Income earned in a week Years of education Number of children

SELECTION EFFECTS

A selection effect occurs in an experiment when the participants in one level of the IV are systematically different than the participants in the other level or levels of the IV. - n an experiment, when the kinds of participants in one level of the independent variable are systematically different from those in the other - can is what happen when you let thme pick their gorup

WITHIN-SUBJECTS DESIGN

A within-subjects design is where the same participants are exposed to different experimental conditions. - Also sometimes called a repeated-measures design. =is a type of within-groups design in which participants are measured on a dependent variable more than once, after exposure to each level of the independent variable. = In addition, the real participant tasted one chocolate at the same time the confederate was also tasting it, but she tasted the other chocolate while the confederate was viewing a painting. The participant was told that the two chocolates were different, but in fact they were exactly the same -\= It was a repeated-measures design because each participant rated the chocolate twice

MODERATION

ARE THERE SUBGROUPS WHERE THE RELATIONSHIP IS STRONGER OR WEAKER? - a moderation hypothesis could propose that the link between conscientiousness and good health is strongest among older people (whose health may be more vulnerable to neglecting doctor's orders) and weakest among younger people (who might stay healthy even if they ignore doctor's orders)- - he word moderate can mean "to change," and a moderating variable can change the relationship between the other two variables (making it more intense or less intense).

INDEPENDENT-GROUPS DESIGNS

Although the minimum requirement for an experiment is that researchers manipulate one variable and measure another, experiments can take many forms. One of the most basic distinctions is between independent-groups designs and within-groups designs. independent-groups design An experimental design in which different groups of participants are exposed to different levels of the independent variable, such that each participant experiences only one level of the independent variable. Also called between-subjects design, ===: A design in which different participants are exposed to different experimental conditions. Each participant experiences only one level of the independent variable. independet -// Between - there in no inbetween = you eather get it or you dont

Ceiling and floor effects

An experimental design problem in which independent variable groups score almost the same on a dependent variable, such that all scores fall at the high end of their possible distribution. = For example, if the researchers really did manipulate the independent variable by giving people $0.00, $0.25, or $1.00, that would be a floor effect because these three amounts are all low—they're squeezed close to a floor of $0.00.

Control Variables

Any variable that an experimenter holds constant on purpose is called a control variable. - Therefore, besides the independent variable, researchers also control potential third variables (or nuisance variables) in their studies by holding all other factors constant between the levels of the independent variable

MEDIATION—

As the name implies, the mediating variable comes in the middle of the other two variables.

SELECTION EFFECTS— MATCHED GROUP

Avoiding Selection Effects with Matched Groups. Some researchers prefer to use matched groups with small samples. ,researchers may wish to be absolutely sure the experimental groups are as equal as possible before they administer the independent variable. In these cases, they may choose to use matched groups, or matching. ==== experimental design technique in which participants who are similar on some measured variable are grouped into sets; the members of each matched set are then randomly assigned to different experimental conditions. ===== ex --- Student achievement, operationalized by GPA, for instance, might matter in the study of notetaking. The researchers would next match up participants in pairs, starting with the two having the highest GPAs, and within that matched set, randomly assign one of them to each of the two notetaking conditions. They would then take the pair with the next-highest GPAs and within that set again assign randomly to the two groups. They would continue this process until they reach the participants with the lowest GPAs and assign them at random too - This method also ensures that the groups are equal on some important variable, such as GPA, before the manipulation of the independent variabl = requires more time and often more resources than random assignment.

GRAPHING ASSOCIATIONS WHEN ONE VARIABLE IS CATEGORICAL

But is a scatterplot the best representation of an association in which one of the variables is measured categorically? - researchers may also plot the results of an association with a categorical variable as a bar graph, as in Figure 8.4. Each person is not represented by one data point; instead, the graph shows the mean marital satisfaction rating (the arithmetic average) for all the people who met their spouses online and the mean marital satisfaction rating for those who met their spouses in person. ANALYZING ASSOCIATIONS WHEN ONE VARIABLE IS CATEGORICAL When at least one of the variables in an association claim is categorical, as in the online dating example, researchers may use different statistics to analyze the data. Although they occasionally use r, it is more common to estimate the magnitude of difference between means (group averages).

IS THE ASSOCIATION CURVILINEAR?

CURVILINEAR = in which the relationship between two variables is not a straight line; it might be positive up to a point and then become negative. - Sometimes a zero correlation, could be curvilinear - Correlations can only determine linear relationships - A curvilinear association exists, for example, between age and the use of health care service = When we compute a simple bivariate correlation coefficient r on these data, we get only r = −.01 because r is designed to describe the slope of the best-fitting straight line through the scatterplot.

Internal Validity:

Can We Make a Causal Inference from an Association? - Even though it's not necessary to formally interrogate internal validity for an association claim, we must guard against the powerful temptation to make a causal inference from any association claim we read.

INTERROGATING CAUSAL CLAIMS WITH THE FOUR VALIDITIES

Construct validity: How well were the variables measured and manipulated? External validity: To whom or what can the causal claim generalize? Statistical validity: How much? How precise? What else is known? Internal validity: Are there alternative explanations for the results?

THREATS EVEN TRUE EXPERIMENTS MAY NOT ELIMINATE

Contamination: communication of information about the experiment between groups Resentment, rivalry, diffusion of treatment Experimenter expectancy effects: experimenter unintentionally influences results Novelty effects Disruption effects

REVIEWING THE THREE CAUSAL CRITERIA

Covariance Temporal precedence Internal validity Multivariate designs -involve more than two measured variables. - longitudinal designs, multiple-regression designs, and the pattern and parsimony approach are Longitudinal designs help address temporal precedence. Multiple regression analyses help address internal validity. The bivariate examples in Chapter 8 involved only two measured variable

APPLYING THE THREE CAUSAL CRITERIA

Covariance: There must be an association between the cause variable (A) and the effect variable (B). = for causal = Based on the results of five studies, we already know deep talk is associated positively with well-being. As the percentage of deep talk goes up, well-being goes up, thus showing covariance of the proposed cause and the proposed effect. Temporal precedence (the directionality problem): The causal variable (A) must come before the effect variable (B). ---- The temporal precedence criterion is sometimes called the directionality problem because we don't know which variable came first. Internal validity (third-variable problem): Is there a third variable (C) that is associated with variables A and B independently? If so, then we can't infer causation. ===== -The internal validity criterion is often called the third-variable problem: When we can come up with an alternative explanation for the association between two variables Spurious association A bivariate association that is attributable only to systematic mean differences on subgroups within the sample; the original association is not present within the subgroup

CROSS-SECTIONAL CORRELATIONS

Cross-sectional correlations: tests of whether two variables measured at the same point in time are correlated - For example, the study reports that the correlation between mothers' overvaluation at Time 4 and children's narcissism at Time 4 was r = .099. This correlation is consistent with the hypothesis. However, because both variables in a cross-sectional correlation were measured at the same time, this result alone cannot establish temporal precedence

WITHIN-GROUPS DESIGNS

Do within-groups designs fulfill the three causal criteria? Covariance: There is a manipulated IV and comparison conditions. Temporal precedence: The IV occurs before the DV. Internal validity: Are there third-variable explanations? Order effects are a potential confound

HOW STRONG IS THE RELATIONSHIP?

Effect size: describes the strength of an association EVALUATING EFFECT SIZE All else being equal, larger effect sizes are more important "Small" effect sizes can compound over many observations - The first thing to understand is that effect sizes can indicate the importance of a result. When all else is equal, a larger effect size is often considered more important than a small one ---- you need to comapre it to what! Benchmarks: "compared to what? "Small" Effect Sizes Can Compound Over Many Observations. Tiny effect sizes can become important when aggregated over many situations, and they can also become important when aggregated over many people.

BIVARIATE CORRELATIONS

Even though each study measured more than two variables, an analysis of bivariate correlations looks at only two variables at a time. Therefore, a correlational study might have measured multiple variables, but the authors present the bivariate correlations between different pairs of variables separately.

EXPERIMENTS CAN SUPPORT CAUSAL CLAIMS

Experiment: Examination of causal relationship between variables using an independent variable = In psychological science, however, = specifically means that the researchers manipulated at least one variable and measured another Variable: A behavior, situation, or characteristic, etc. that can differ from individual to individual Independent variable (a.k.a. factor): Variable that is manipulated making multiple conditions (a.k.a. levels) in an experiment - temp in class ======== =the researcher has some "independence" in assigning people to different levels of this variable A manipulated variable is a variable that is controlled, such as when the researchers assign participants to a particular level (value) of the variable. condition One of the levels of the independent variable in an experiment. Dependent variable/outcome variable -What the researcher measures/observes/ gathers data from the individual ======= Measured variables take the form of records of behavior or attitudes, such as self-reports, behavioral observations, or physiological measures

CONSTRUCT VALIDITY AND THEORY TESTING

Experiments are designed to test theories. Therefore, interrogating the construct validity of an experiment requires you to evaluate how well the measures and manipulations researchers used in their study capture the conceptual variables in their theory

partial counterbalancing

From a complete counterbalancing, choose a subset or eliminate sequences that will confound the interpretation of results.- Choose the sequences that meet the criteria to control for order and carryover effects a method of counterbalancing in which some, but not all, of the possible condition orders are represented - in which only some of the possible condition orders are represented. One way to partially counterbalance is to present the conditions in a randomized order for every subject. =Another technique for partial counter balancing is to use a Latin square, a formal system to ensure that every condition appears in each position at least once.

Full counterbalancing / COMPLETE COUNTERBALANCING

Full counterbalancing A method of counterbalancing in which all possible condition orders are represented. - . When a within-groups experiment has only two or three levels of an independent variable, researchers can use - A simple example with two conditions (treatment and control): 50% of participants experience control then treatment 50% of participants experience treatment then control - When you have more than two conditions use must use a factorial, N! N! where N is the number of conditions N! = N x N-1 x N-2..... X 2 x 1 3 = N 3X2X1 = 6 Each condition must occur equally often = design with two conditions is easy to counterbalance because there are only two orders (A → B and B → A). In a repeated-measures design with three conditions—A, B, and C—each group of participants could be randomly assigned to one of the six following sequences: For every possible order, there needs to be at least one participant which is easy to achieve if there is 2, 3, or 4 conditions.

EXTERNAL VALIDITY: TO WHOM OR WHAT CAN THE CAUSAL CLAIM GENERALIZE?

Generalizing to other people . Remember that when you interrogate external validity, you ask about random sampling—randomly gathering a sample from a population. - (In contrast, when you interrogate internal validity, you ask about random assignment—randomly assigning each participant in a sample into one experimental group or another.) Generalizing to other situations - External validity also applies to the types of situations to which an experiment might generalize. For example, the notetaking study used five videotaped TED talk lectures eported two additional experiments, each of which used new video lectures. All three experiments found the same pattern, so you can infer that the effect of laptop notetaking does generalize to other TED talks To decide whether an experiment's results can generalize to other situations, we need to conduct more research What if external validity is poor? Remember from Chapter 3 that in an experiment, researchers usually prioritize experimental control—that is, internal validity. To get a clean, confound-free manipulation, they may have to conduct their study in an artificial environment like a university laboratory. Such locations may not represent situations in the real world. Although it's possible to achieve both internal and external validity in a single study, doing so can be difficult. Therefore, many experimenters decide to sacrifice real-world representativeness for internal validity.

STATISTICAL VALIDITY: /significantce :HOW MUCH? HOW PRECISE? WHAT ELSE IS KNOWN?

How large is the effect? - n Mueller and Oppenheimer's first study (2014), the effect size for the difference in conceptual test performance between the longhand and laptop groups was d = 0.38. This means the laptop group scored 0.38 of a standard deviation higher than the longhand group. Psychologists sometimes start by saying a d of 0.2 should be considered small, a d of 0.5 is moderate, and a d of 0.8 is large ( -0 -you learned that the correlation coefficient r helps researchers evaluate the effect size (strength) of an association How precise is the estimate? (95% CI) are computed so that 95% of them will contain the true population difference. - The width of the 95% CI reflects precision. When a study has a relatively small sample and more variability in the data, the CI will be relatively wide (less precise). When a study has a larger sample and less variability, then the CI will be narrower (more precise) Has the study been replicated?

STATISTICAL VALIDITY

How strong is the relationship? How precise is the estimate? Has it been replicated? Could outliers be affecting the association? Is there restriction of range? Is the association curvilinear?

CHAPTER 10:

INTRODUCTION TO SIMPLE EXPERIMENTS

CONSTRUCT VALIDITY: HOW WELL WERE THE VARIABLES MEASURED AND MANIPULATED?

In an experiment, researchers operationalize two constructs: the independent variable and the dependent variable. - When you interrogate the construct validity of an experiment, you should ask about the construct validity of each of these variables. DEPENDENT VARIABLES: HOW WELL WERE THEY MEASURED? 2. Independent variables: How well were they manipulated? Manipulation checks =is an extra dependent variable that researchers can insert into an experiment to convince them that their experimental manipulation worked/ an extra dependent variable researchers can include to determine how well a manipulation worked. === interested in investigating whether humor would improve students' memory of a college lecture (Kaplan & Pascoe, 1977). Students were randomly assigned to listen to a serious lecture or one punctuated by humorous examples, and the key dependent variable was their memory for the material. In addition, to ensure they actually found the humorous lecture funnier than the serious one, students rated the lecture on how "funny" and "light" it was. Pilot studies =A study completed before (or sometimes after) the study of primary interest, usually to test the effectiveness or characteristics of the manipulations = is a simple study, using a separate group of participants =might have exposed a separate group of students to either a serious or a humorous lecture and then asked them how amusing they found it, 3. Construct validity and theory testing

WHY NOT JUST DO AN EXPERIMENT?

In many cases participants cannot be randomly assigned to a variable. People cannot be assigned to preferences. It may be unethical to assign participants.

posttest-only design

In this design, participants are randomly assigned to independent variable groups and are tested on the dependent variable once - The notetaking study is an example of a posttest-only design, with two independent variable levels --Participants were randomly assigned to a laptop condition or a longhand condition and they were tested only once on the video they watched. - Posttest-only designs satisfy all three criteria for causation. They allow researchers to test for covariance by detecting differences in the dependent variable They establish temporal precedence because the independent variable comes first in time. And when they are conducted well, they establish internal validity.

EXPERIMENTS ESTABLISH COVARIANCE

Independent variables answer the question, "Compared to what?" Comparison group (comparison condition) =A group in an experiment whose levels on the independent variable differ from those of the treatment group in some intended and meaningful way. Also called comparison condition. Covariance: it is also about the results There are a couple of ways an independent variable might be designed to show covariance Control groups, treatment groups, and comparison groups Control group (no treatment condition) Treatment group(s) (one or more treatment conditions) Placebo group (placebo control group) There are a couple of ways an independent variable might be designed to show covariance

WELL-DESIGNED EXPERIMENTS ESTABLISH INTERNAL VALIDITY

Internal validity is subject to a number of distinct threats, three of which—design confounds, selection effects, and order effects A well-designed experiment establishes internal validity, which is one of the most important validities to interrogate when you encounter causal claims. Design confounds Selection effects Confound (confuse) =For any given research question, there can be several possible alternative explanations, which are known as confounds, or potential threats to internal validity. =The word confound can mean "confuse": When a study has a confound, you are confused about what is causing the change in the dependent variable - Before treatment, group performance needs to be equal.

GETTING AT CAUSALITY WITH PATTERN AND PARSIMONY

Longitudinal correlational designs can satisfy the temporal precedence criterion. - Multiple-regression analyses statistically control for some potential internal validity problems (third variables). - In this section, we explore how researchers can investigate causality by using a variety of correlational studies that all point in a single, causal direction ===== This approach can be called "pattern and parsimony" because there's a pattern of results best explained by a single, parsimonious causal theory ==== Parsimony: simplicity; the degree to which a good scientific theory provides the simplest explanation of some phenomenon = means the simplest explanation of a pattern of data—the theory that requires making the fewest exceptions or qualifications. - The Power of Pattern and Parsimony 1.The longer a person has smoked cigarettes, the greater are the chances of getting cancer. 2.People who stop smoking have lower cancer rates than people who continue smoking. 3.Smokers' cancers tend to be in the lungs and of a particular type. 4.Smokers who use filtered cigarettes have a somewhat lower rate of cancer than smokers of unfiltered cigarettes. 5.People who live with smokers will also have higher rates of cancer because of their passive exposure to the same chemicals.

LONGITUDINAL DESIGNS—ESTABLISHING TEMPORAL PRECEDENCE

Longitudinal design: Can provide evidence for temporal precedence by measuring the same variables in the same people at several different times Used by developmental psychologists Gets us closer to a causal claim Interpreting Results from Longitudinal Designs =Because there are more than two variables involved, a multivariate design gives several individual correlations, referred to as cross-sectional correlations, autocorrelations, and cross-lag correlations

LONGITUDINAL STUDIES AND THE THREE CRITERIA FOR CAUSATION

Longitudinal designs can provide some evidence for causation by fulfilling three criteria Covariance -satistical relationships in longitudinal designs help establish covariance. When two variables are correlated and their 95% CIs do not contain zero (as in the cross-lag correlations in Figure 9.3), there is covariance. Temporal precedence -have an idea of came first Internal validity -When conducted simply—by measuring only the two key variables—longitudinal studies may not help rule out third variables. For example, the Brummelman results presented in Figure 9.3 cannot clearly rule out the possible third variable of socioeconomic status. It's possible that parents in higher income brackets overpraise their children, and that children in upper-income families are also more likely to think they're better than other kids. = However, researchers can sometimes design their studies or conduct subsequent analyses in ways that address some third variables

CHAPTER 9:

MULTIVARIATE CORRELATIONAL RESEARCH - btw - assassiion claims lay the grown work for causal claims == thats why covarince is one of the crittiria to establish causal claims

MEDIATORS VERSUS THIRD VARIABLES

Mediators are similar to third-variable explanations because both of them can be tested with multiple regression. However, they tell different theoretical stories about a relationship. Similarities Both involve multivariate research designs. Both can be detected using multiple regression. Differences Third variables are external to the bivariate correlation (problematic). = external to the two variables in the original bivariate correlation.= It might even be seen as an accident—a problematic "lurking variable" that potentially distracts from the relationship of interest. - For example, if we propose that education level is a third variable responsible for the deep talk/well-being relationship, we're saying deep talk and well-being are correlated with each other only because each one is correlated separately with education, Mediators are internal to the causal variable (not problematic). - In contrast, a mediation hypothesis tells a theoretically meaningful, step-by-step story in which "A leads to M leads to B" (for example, deep talk leads to social ties, which lead to well-being). A mediator variable is of direct interest to the researchers rather than a nuisance. btw Because mediation hypotheses are causal claims, mediation is definitively established only in conjunction with temporal precedence

MEDIATORS VERSUS MODERATORS

Mediators: Why are these two variables linked? To mediate means to come in the middle of the other two variables. Moderators: Are these two variables linked in the same way for all participants or in every situation?

THREATS TO INTERNAL VALIDITY

Non-equivalent groups/Selection effects' Occurs when differences exist between individuals in treatment and control groups at the start of a study These differences become alternative explanations for any differences observed at the end of the study Random assignment controls the selection threat Suppose a community recycling program is tested. Individuals who are interested in recycling are encouraged to participate

Three Potential Internal Validity Threats in Any Study

OBSERVER BIAS - Experimenter/observer bias: Occurs when a researcher inadvertently treats groups differently in the study due to knowledge of the hypothesis for the study. = when researchers' expectations influence their interpretation of the results. Social desirability: Participants present themselves in a more desirable way Demand characteristics: Participants try to "figure out" the study and change behaviors to match expectations Diffusion of treatment: When subjects in one condition share information about the study with subjects in the other condition For the most part, using single-blind or double-blind procedure will address these issues = . The most appropriate way to avoid such problems is to conduct a double-blind study, in which neither the participants nor the researchers who evaluate them know who is in the treatment group and who is in the comparison group. - When a double-blind study is not possible, a variation might be an acceptable alternative. In some studies, participants know which group they are in, but the observers do not; this is called a masked design, or blind design

- 1. Explain why experiments are superior to multiple-regression designs for controlling for third variables.

One problem is that even though multivariate designs analyzed with regression statistics can control for third variables, they cannot always establish temporal precedence. - Of course, the Chandra study did measure viewing sexual TV content 3 years before pregnancies occurred. But others, such as the study on family meals and academic achievement, may not. - Even when a study takes place over time (longitudinally), another very important problem is that researchers cannot control for variables they do not measure. Even though multiple regression controls for any third variables the researchers do measure in the study, some other variable they did not consider could account for the association. The "lurking-variable," or third-variable problem is one reason a well-run experimental study is ultimately more convincing in establishing causation than a correlational study. - An experimental study on TV, for example, would randomly assign a sample of people to watch either sexy TV shows or programs without sexual content. The power of random assignment would make the two groups likely to be equal on any third variables the researchers did not happen to measure, such as religiosity, social class, or parenting styles. -But of course, just like randomly assigning children to get one type of praise or another, it is ethically questionable to conduct an experiment on sexual TV content. - A randomized experiment is the gold standard for determining causation. Multiple regression, in contrast, allows researchers to control for potential third variables, but only for the variables they choose to measure.

Carryover effects

Order effects also include carryover effects, = in which some form of contamination carries over from one condition to the next. imagine sipping orange juice right after brushing your teeth; the first taste contaminates your experience of the second one. = ex= An order effect in the chocolate-tasting study could have occurred if people rated the first chocolate higher than the second simply because the first bite of chocolate is always the best; subsequent bites are never quite as good

STATISTICAL VALIDITY QUESTION 5: IS THERE RESTRICTION OF RANGE?

RESTRICTION OF RANGE -A situation involving a bivariate correlation, in which there is not a full range of possible scores on one of the variables in the association, so the relationship from the sample underestimates the true correlation. = , if there is not a full range of scores on one of the variables in the association, it can make the correlation appear smaller than it really is - Because restriction of range usually makes correlations appear smaller, we would ask about it primarily when the correlation appears weaker than expected.

INTERPRETING CORRELATIONS— OUTLIERS

STATISTICAL VALIDITY QUESTION 4: COULD OUTLIERS BE AFFECTING THE ASSOCIATION? An outlier is an extremely deviant individual in the sample that is characterized by a much larger (or smaller) score than all the others in the sample. In a scatter plot, the point is clearly different from all the other points. Outliers produce a disproportionately large impact on the correlation coefficient.

SELECTED ORDERS

Selected orders is an incomplete/partial counterbalancing technique that is used when the IV has 4 or more levels. With this approach, the researcher selects a particular set of orders of conditions to balance practice effects.

MEDIATORS VERSUS THIRD VARIABLES

Start with a basic correlation between A and B. A moderator is kind of interesting—it means your correlation is there in one group, but not the other. However, a third variable is a nuisance or a problem: if you want to make a claim that A and B are correlated, a plausible third variable would mean that your correlation isn't really "there" at all.

RANDOM STARTING ORDER WITH ROTATION

Start with a random order, and then slowly rotate between the orders. - A counterbalancing technique that controls for to control for order effects without using all possible orders, but the order of conditions is not balanced.

EXPERIMENTS ESTABLISH TEMPORAL PRECEDENCE

The cause variable precedes the effect variable - By manipulating the independent variable, the experimenter virtually ensures that the cause comes before the effect (or outcome). The ability to establish temporal precedence is a feature that makes experiments superior to correlational designs. A simple correlational study is a snapshot—all variables are measured at the same time

CORRELATION EXAMPLE

The number of weeks each of 10 patients took a bipolar medication and their Beck Depression Inventory-II (BDI-II) score at the end of the medication period is presented in the table below. corration reseaach is the begging of experimental reseach -

WHICH DESIGN IS BETTER?

The research situation determines which design is better. In some situations, it is problematic to use a pretest/posttest design. In other situations, it makes sense to use a pretest/posttest design. Posttest-only designs can nonetheless be very powerful.

d

The second way is to use a standardized effect size. In Chapter 8, you learned that the correlation coefficient r helps researchers evaluate the effect size (strength) of an association. When there are two groups in an experiment, we often use an indicator called d. This standardized effect size takes into account both the difference between means and the spread of scores within each group (the standard deviation). When d is large, it means the independent variable caused a large change in the dependent variable, relative to how spread out the scores are. When d is small, it means the scores of participants in the two experimental groups overlap more. - Psychologists sometimes start by saying a d of 0.2 should be considered small, a d of 0.5 is moderate, and a d of 0.8 is large (

Three Possible Patterns from a Cross-Lag Study.

The study did show that parental overpraise (overvaluation) at earlier times was correlated with child narcissism at the later times. Such a pattern was consistent with the argument that overpraise leads to increases in narcissism over time. However, the study could have shown the opposite result—that narcissism at earlier times was correlated with overpraise later. Such a pattern would have indicated that the childhood narcissistic tendency came first, leading parents to change their type of praise later. Finally, the study could have shown that both correlations were different from zero—that overpraise at Time 1 predicted narcissism at Time 2 and that narcissism at Time 1 predicted overpraise at Time 2. If that had been the result, it would mean excessive praise and narcissistic tendencies are mutually reinforcing. In other words, there is a cycle in which overpraise leads to narcissism, which leads parents to overpraise, and so on.

effect size

The term effect size describes the strength of a relationship between two or more variables

INTERROGATING CAUSAL CLAIMS

There are three criteria for causation: 1. Covariance: This simply means that the two variables are related. 2. Temporal precedence: one variable comes before the other variable in time. 3. Internal validity (also known as the thirdvariable criterion): a study should be able to eliminate alternative explanation WHY EXPERIMENTS SUPPORT CAUSAL CLAIMS = Experiments Establish Covariance -In this case, covariance is indicated by a difference in the group means:

STATISTICAL VALIDITY QUESTION 2: HOW PRECISE IS THE ESTIMATE?

To communicate the precision of their estimate of r, researchers report a 95% confidence interval (95% CI), == , but we do know that CIs are designed to capture the true relationship in 95% of studies like this. Sample size and precision ============================ -When an estimate is based on a small sample, it is less stable. -Reflecting this instability, small samples have wider (less precise) confidence intervals. The CI has to be wide to capture the degree of uncertainty we have when the sample is smal - -large samples result in estimates with much narrower, more precise confidence intervals. Confidence Intervals that do not contain zero ============================= . When the 95% CI does not include zero, it is common to say that the association is statistically significant. The definition of a statistically significant correlation is one that is unlikely to have come from a population in which the association is zero. Confidence Intervals that do contain zero 0 we can't rule out that the true association is zero.- - When the 95% CI contains zero, it is common to say that the association is "not statistically significant."

USING STATISTICS TO CONTROL FOR THIRD VARIABLES

To control for means to hold constant. When conducting a multivariate design, the relationship between two variables still exists when controlling for another variable.

PERHAPS THERE IS NOT ENOUGH BETWEEN-GROUPS DIFFERENCE

Weak manipulations = What if the amounts were $0.00, $0.25, and $1.00? In that case, it might be no surprise that the manipulation didn't have a strong effect. A dollar doesn't seem like enough money to affect most people's mood Insensitive measures = Sometimes a study finds a null result because the researchers have not used an operationalization of the dependent variable with enough sensitivity. If a medication reduces fever by a tenth of a degree, you wouldn't be able to detect it with a thermometer that was calibrated in one-degree increments Ceiling and floor effects Manipulation checks to help detect weak manipulations, ceilings, and floors Design confound acting in reverse

Pattern, Parsimony, and the Popular Media

When journalists write about science, they do not always fairly represent pattern and parsimony in research. - When journalists report only one study at a time, they selectively present only a part of the scientific process. They might not describe the context of the research, such as what previous studies have revealed or what theory the study was testing.- Journalists do not always fairly represent pattern and parsimony. When journalists report only one study at a time, they are selectively presenting only part of the scientific process.

MODERATING VARIABLES

When the relationship between two variables changes depending on the level of another variable, that other variable is called a moderator.

MANIPULATION CHECKS HELP DETECT WEAK MANIPULATIONS, CEILINGS, AND FLOORS

When you interrogate a study with a null effect, it is important to ask how the independent and dependent variables were operationalized. 0 manipulation check is a separate dependent variable that experimenters include in a study, specifically to make sure the manipulation worked. - . If the manipulation check worked, the researchers could look for another reason for the null effect of anxiety on logical reasoning

THREATS TO INTERNAL VALIDITY:- -1. design confounds, selection effects, and order effects.

With a design confound, there is an alternative explanation because the experiment was poorly designed; another variable happened to vary systematically along with the intended independent variable - . If the test questions assigned to the laptop condition were more difficult than the those assigned to the pen condition, that would have been a design confound (see Figure 10.5). It would not be clear whether the notetaking format or the difficulty With a selection effect, a confound exists because the different independent variable groups have systematically different types of participants - However, we are not sure if their improvement was caused by the therapy or by greater overall involvement on the part of the parents who elected to be in the intensive-treatment group. Those parents' greater motivation With an order effect (in a within-groups design), there is an alternative explanation because the outcome might be caused by the independent variable, but it might also be caused by the order in which the levels of the variable are presented. When there is an order effect, we do not know whether the independent variable is really having an effect, or whether the participants are just getting tired, bored, or well-practiced.

INTERROGATING ASSOCIATION CLAIMS

With an association claim, the two most important validities to interrogate are construct validity and statistical validity. Construct validity: How well was each variable measured? --- so imma look at those measurmenand ask if it valid and is it reliabile and valib and see it there are legit - Statistical validity: How well do the data support the conclusion? - you are asking about factors that might have affected the scatterplot, correlation coefficient r, bar graph, or difference score that led to your association claim. You need to consider the strength and precision of your estimate, if it has been replicated, any outliers that might have affected the overall findings, restriction of range, and whether a seemingly zero association might actually be curvilinear. - STATISTICAL VALIDITY QUESTION 1: HOW STRONG IS THE RELATIONSHIP? Internal validity: Can we make a causal inference from an association? External validity: To whom can the association be generalized?

THREATS TO INTERNAL VALIDITY

With no comparison group, must rule out: history, maturation, testing, instrumentation, regression, subject attrition, selection When there is a comparison group, you must rule out these threats: selection, differential regression, additive effects with selection Adding a comparison group helps rule out many threats to internal validity.

INTERNAL VALIDITY: CONTROLLING FOR ORDER EFFECTS

Within-groups designs have the potential for a particular threat to internal validity: Order Effects Sometimes, being exposed to one condition first changes how participants react to the later condition. - and they happen when exposure to one level of the independent variable influences responses to the next level = An order effect in a within-groups design is a confound, meaning that behavior at later levels of the independent variable might be caused not by the experimental manipulation but rather by the sequence in which the conditions were experienced.

MATURATION THREATS

a change in behavior that emerges more or less spontaneously over time - People adapt to changed environments; children get better at walking and talking; plants grow taller—but not because of any outside intervention. It just happens. - Perhaps. An alternative explanation, however, is that most of them simply settled in, or "matured into," the camp setting after they got used to the place - Similarly, the depressed women may have improved because the cognitive therapy was effective, but an alternative explanation is that a systematically high portion of them simply improved on their own. Preventing Maturation Threats =also have included an appropriate comparison group. - Dr. Yuki would have studied a comparison group of women who started out equally depressed but did not receive the cognitive therapy. If the treatment groups improved significantly more than the comparison groups did, these researchers could essentially subtract out the effect of maturation when they interpret their results.

CAUSAL CLAIMS

a claim arguing that a specific change in one variable is responsible for influencing the value of another variable - A causal claim argues that one variable causes changes in the level of another variable. - needs to be testable - Causal claims are supported by experiments (studies that have a manipulated variable and a measured variable.

cofounding variable

a factor other than the independent variable that might produce an effect in an experiment

Spurious association

a relationship between a presumed cause and a presumed effect that occurs as a result of an unexamined variable or set of variables - Spurious association

TESTING THREATS

a specific kind of order effect, refers to a change in the participants as a result of taking a test (dependent measure) more than once. People might have become more practiced at taking the test, leading to improved scores, or they may become fatigued or bored, taking test effect Preventing Testing Threats. To avoid testing threats, researchers might abandon a pretest altogether and use a posttest-only design - If they do use a pretest, they might opt to use alternative forms of the test for the two measurements

point estimate

a summary statistic from a sample that is just one number used as an estimate of the population parameter

Novelty effects

a threat to external validity that occurs when the knowledge that what is being done is new and under study somehow affects the outcome, either favorably or unfavorably.

INSTRUMENTATION THREATS

a threat to internal validity that occurs when a measuring instrument changes over time from having been used before - in observational research, the people who are coding behaviors are the measuring instrument, and over a period of time, they might change their standards for judging behavior by becoming stricter or more lenient -= Thus, maybe Nikhil's campers did not really become less disruptive; instead, the people judging the campers' behavior became more tolerant of loud voices and rough-and-tumble p Preventing Instrumentation Threats. -To prevent instrumentation threats, researchers can switch to a posttest-only design, or they can take steps to ensure that the pretest and posttest measures are equivalent. -To do so, they might collect data from each instrument to be sure the two are calibrated the same Observer bias—fatigue, expectations

Measurement error

an error that occurs when there is a difference between the information desired by the researcher and the information provided by the measurement process

floor effects

an experimental design problem in which independent variable groups score almost the same on a dependent variable, such that all scores fall at the low end of their possible distribution = As special cases of weak manipulations and insensitive measures, ceiling and floor effects can cause independent variable groups to score almost the same on the dependent variable.

Design confounds

another variable accidentally varies systematically along with the IV - A threat to internal validity in an experiment in which a second variable happens to vary systematically along with the independent variable and therefore is an alternative explanation for the results. -The accidental second variable is therefore an alternative explanation for the results- = = ex= If the adult models in the baby study had accidentally exhibited more cheerful attitudes in the effort than the no-effort condition, the study would have a design confound because the second variable (model's cheerfulness) would have systematically varied along with the independent variable (effort versus no effort Systematic variability is the problem Systematic variability Unsystematic variability

ATTRITION THREATS TO INTERNAL VALIDITY

attrition = lose strghts = lose participants Attrition can happen when a pretest and posttest are administered on separate days and some participants are not available on the second day. An attrition threat becomes a problem for internal validity when attrition is systematic; that is, when only a certain kind of participant drops out. - If any random camper leaves midweek, it might not be a problem for Nikhil's research, but it is a problem when the most rambunctious camper leaves early - Similarly, the level of depression among Dr. Yuki's patients might have decreased because of the cognitive therapy, but it might have been because three of the most depressed women in the study could not maintain the treatment regimen and dropped out of the study. The posttest average is lower only because these extra-high scores are not included. Preventing Attrition Threats. -When participants drop out of a study, most researchers will remove those participants' scores from the pretest average too.

BENCHMARKS=== STATISTICAL VALIDITY QUESTION 1: HOW STRONG IS THE RELATIONSHIP?

baseline values the system seeks to attain - : COMPARED TO WHAT? A third way to think about effect size is to compare it to well-understood benchmarks. Recall that the average effect size in psychology studies is around r = .20 and may only rarely be as high as r = .40. the size of the sample is correlated to the the stands used - biggers is better

AVOIDING ORDER EFFECTS BY COUNTERBALANCING

counterbalancing they present the levels of the independent variable to participants in different sequences. -With counterbalancing, any order effects should cancel each other out when all the data are combined.- == In a repeated-measures experiment, presenting the levels of the independent variable to participants in different sequences to control for order effects. = Boothby and her colleagues (2014) used counterbalancing in their experiment (Figure 10.17). Half the participants tasted their first chocolate in the shared condition followed by a second chocolate in the unshared condition. The other half tasted chocolate in the unshared followed by the shared condition. Therefore, the potential order effect of "first taste of chocolate" was present for half of the people in each conditio Two types of counterbalancing - Full counterbalancing - Partial counterbalancing (example: Latin square)

criterion variable and predictor variable

criterion variable The variable in a multiple-regression analysis that the researchers are most interested in understanding or predicting. Also called dependent variable. = criterion = critical = importnat = saute after = Dependent = measure predictor variable A variable in multiple-regression analysis that is used to explain variance in the criterion variable. Also called independent variable. preditcs the uutcome - influces outcome - independent varibel

dichotomous

divided or dividing into two parts or classifications

Pretest/Posttest Design

equivalent groups, pretest/posttest design, participants are randomly assigned to at least two groups and are tested on the key dependent variable twice—once before and once after exposure to the independent variable - A study on the effects of mindfulness training, introduced in Chapter 1, is an example of a pretest/posttest design. In this study, 48 students were randomly assigned to participate in either a 2-week mindfulness class or a 2-week nutrition class (Mrazek et al., 2013). One week before starting their respective classes, all students completed a verbal-reasoning section of a GRE test. One week after their classes ended, all students completed another verbal-reasoning GRE test of the same difficulty. The results, shown in Figure 10.13, revealed that, while the nutrition group did not improve significantly from pretest to posttest, the mindfulness group scored significantly higher at posttest than at pretest. - Researchers might use a pretest/posttest design when they want to be sure random assignment made groups equal

SCALES OF MEASUREMENT

every data we have have a Scale of measurement - tell us what we can learn from that data Categorical versus Quantitative variables!!! Categorical variables (nominal variables) -raw frequency counts Also called nominal variable. Quantitative variables Types of quantitative variables Ordinal scale Interval scale Ratio scale

MODERATING VARIABLES

factors that affect the relationship between the independent and dependent variables] = When the relationship between two variables changes depending on the level of another variable, that other variable is called a moderator. = ex = correlation between professional sports games attendance and the success of the team.= Oishi and his team determined that in cities with high residential mobility, there is a positive correlation between success and attendance, indicating a fair-weather fan base. In cities with low residential mobility, there is not a strong correlation between success and attendance. We say that the degree of residential mobility moderates the relationship between success and attendance (Table

unsystematic variability

in an experiment, when levels of a variable fluctuate independently of experimental group membership, contributing to variability within groups

MEDIATION

he mediator acts as a communicating agent between the parties and suggests ways in which the parties can resolve their dispute. Once researchers have established a relationship between two variables--, they often want to explore it further by thinking about why. =why = mediator A variable that helps explain the relationship between two other variables. Also called mediating variable.- - ex = We know conscientious people are more physically healthy than less conscientious people. But why? The mediator of this relationship might be the fact that conscientious people are more likely to follow medical advice and instructions, and that's why they're healthier. Start with an association between two variables, A and B. = Mediation hypotheses propose a causal mechanism for a bivariate relationship. Why are these two variables correlated? Mediation hypotheses are causal statements. Mediators specify a time sequence for the three variables (temporal precedence). Mediators also specify the mechanism.

SELECTING A HYPOTHESIS

hypothesis comes from scientists simply observing what is going on in the environment and then testing that observation.

external validity of an association claim

interrogating the external validity of an association claim, you ask whether the association can generalize to other people, places, and times. - the first questions would be who the participants were and how they were selected. - As you interrogate external validity, recall that the size of the sample does not matter as much as the way the sample was selected from the population of interest - If the adults were not chosen by a random sample of the population of interest, you could not be sure the sample's results would generalize to that population.

Describing Associations Between Two Quantitative Variables

next step in testing an association claim is to describe the relationship between the two measured variables using scatterplots and the correlation coefficient r. - positive r means that the relationship is positive: - . Across many areas of psychology, correlations are typically around r = .20, but some areas of psychology might average as high as r = .40

REGRESSION RESULTS INDICATE WHETHER A THIRD VARIABLE EXPLAINS THE RELATIONSHIP

no longer a prob bc you are controlling for it - Criterion variable: = dependent variable Predictor variables: independent variables Beta is similar to r. Direction (positive or negative) and strength (the closer to -1 or +1, -the stronger the relationship; the closer to zero, the weaker the relationship). Beta is standardized; b is unstandardized = "in the raw units" 95% CIs and statistical significance of beta

INTERROGATING NULL EFFECTS: WHAT IF THE INDEPENDENT VARIABLE DOES NOT MAKE A DIFFERENCE?

null effect A finding that an independent variable did not make a difference in the dependent variable; there is no significant covariance between the two. Also called null result. - Here are three hypothetical examples: Any time an experiment gives a null result, it might be the case that the independent variable has virtually no effect on the dependent variable. In the real world, perhaps money does not make people any happier, - A different possibility is that there is a true effect, but this particular study did not detect it

Order Effects

occur when the order in which the participants experience conditions in an experiment affects the results of the study Practice effects or Carryover effects

CONCURRENT-MEASURES DESIGN

participants are exposed to all the levels of an independent variable at roughly the same time, and a single attitudinal or behavioral preference is the dependent variable. - infant cognition, in which infants were shown a male face and a female face at the same time, and an experimenter recorded which face they looked at the longest (Quinn et al., 2002). The independent variable was the gender of the face, and babies experienced both levels (male and female) at the same time. The baby's looking preference was the dependent variable (Figure 10.15). This study found that babies show a preference for looking at female faces, unless their primary caretaker is male.

ALTERNATIVES TO THE PEARSON CORRELATION

pearson wa devolved for Linear relationships - but others have been developed for non-linera data - Spearman's rho (ρ): Coefficient is used when at least one variable is measured on an ordinal scale (rank-order data) - Also may be used if with interval/ ratio data, when the relationship is consistently directional but may not be linear. Point-Biserial Correlation: Coefficient is used when one variable is interval/ ratio and one variable is dichotomous/ nominal. Phi (ϕ) Coefficient: Coefficient is used when both variables are dichotomous and/or nominal.

Two basic forms of independent-groups designs

posttest-only design and - the pretest/posttest design. - The two types of designs are used in different situations.

Regression Results Indicate Whether a Third Variable Affects the Relationship

predictor variables, or independent variables. In the sexy TV/pregnancy study, the predictor variables are the amount of sexual content teenagers reported viewing on TV and the age of each teen - USING BETA TO TEST FOR THIRD VARIABLES Beta Basics there is often a column labeled beta (or β, or even standardized beta). - There will be one beta value for each predictor variable. Beta is similar to r, but it reveals more than r does. A positive beta, like a positive r, indicates a positive relationship between that predictor variable and the criterion variable, when the other predictor variables are statistically controlled for. -A negative beta, like a negative r, indicates a negative relationship between two variables (when the other predictors are controlled for) - A beta that is zero, or nearly zero, represents no relationship (when the other predictors are controlled for) include the symbol b instead of beta. The coefficient b represents an unstandardized coefficient. A b is similar to beta in that the sign of b—positive or negative—denotes a positive or negative association (when the other predictors are controlled for). - But unlike two betas, we cannot compare two b values within the same table to each other. - The reason is that b values are computed from the original measurements of the predictor variables (such as dollars, centimeters, or inches), whereas betas are computed from predictor variables that have been changed to standardized units

REGRESSION THREATS

refers to a statistical concept called regression to the mean. -When a group average (mean) is unusually extreme at Time 1, the next time that group is measured (Time 2), it is likely to be less extreme—closer to its typical or average performance. - Regression threats occur only when a group is measured twice, and only when the group has an extreme score at pretest. If the group has been selected because of its unusually high or low group mean at pretest, you can expect them to regress toward the mean somewhat when it comes time for the posttes Everyday Regression to the Mean. during an early round of the 2019 Women's World Cup, the team from Italy outscored the team from Jamaica 5-0. That's a big score; soccer (football) teams hardly ever score 5 points in a game. Without being familiar with either team, people who know about soccer would predict that in their next game, Italy would score fewer than 5 goals. Why? Simply because most people have an intuitive understanding of regression to the mean. = Here's another example. Suppose you're normally cheerful and happy, but on any given day your usual upbeat mood can be affected by random factors Preventing Regression Threats. Once again, comparison groups can help researchers prevent regression threats, along with a careful inspection of the pattern of results

Selected orders

select particular orders of conditions to balance practice effects - Selected orders is an incomplete/partial counterbalancing technique that is used when the IV has 4 or more levels. - With this approach, the researcher selects a particular set of orders of conditions to balance practice effects. Each level of the IV appears in each ordinal position only once. The number of selected orders must be equal to a multiple of the number of conditions in the experience.

CROSS-LAG CORRELATIONS

show whether the earlier measure of one variable is associated with the later measure of the other variable - t(the meat of it) Cross-lag correlations: Correlations of the degree to which an earlier measure of one variable is associated with a later measure of the other variable Examines how people change over time. Researchers are most interested in cross-lag correlationsbecause they help establish temporal precedence. - - By inspecting the cross-lag correlations in a longitudinal design, we can investigate how one variable correlates with another one (that's the "cross" part of its name) over time (that's the "lag" part)

95% CIs and Statistical Significance of Beta.

sig or p, or may have an asterisked footnote giving a p value for each beta. -Recall that a p value of .05 complements the .95 from a 95% CI. - Specifically, when the p value is less than .05, you can infer that the 95% CI for that beta does not contain zero and is therefore considered statistically significant. = p < .05, or statistically significant. When p is greater than .05, the beta is considered not significant (n.s.), and you can infer that its 95% CI does contain zero "The 95% CI for the relationship between exposure to sex on TV and pregnancy does not contain zero, suggesting that this relationship is positive, controlling for age." - *p < .05, meaning the result is statistically significant and the 95% CI does not include zero.

CONSTRUCT VALIDITY

the extent to which variables measure what they are supposed to measure - Ask about the construct validity of each variable. How well was each of the variables measured? Does the measure have good reliability? Is it measuring what it's intended to measure? What is the evidence for its face validity, for its concurrent validity, and for its discriminant and convergent validity? An association claim describes the relationship between two measured variables, so it is relevant to ask about the construct validity of each variable. - To interrogate the Mehl study, for example, you would ask questions about the researchers' operationalizations of deep talk and well-being

Situation noise

unrelated events or distractions in the external environment that create unsystematic variability within groups in an experiment

HISTORY THREATS

which result from a "historical" or external factor that systematically affects most members of the treatment group at the same time as the treatment itself, making it unclear whether the change is caused by the treatment received. -To be a history threat, the external factor must affect most people in the group in the same direction (systematically), not just a few people (unsystematically). In the third example, why did the dorm residents use less electricity? Was it the Go Green campaign? Perhaps. But a plausible alternative explanation is that the weather got cooler and most residents did not use air conditioning as much Preventing History Threats As with maturation threats, a comparison group can help control for history threats. - (This would be a pretest/posttest design rather than a one-group, pretest/posttest design.) If both groups decreased their electricity usage about the same over time