PSY3100 Ch10

When interrogating the construct validity of the dependent variable in an experiment, which of the following questions should be asked? "How well was this variable manipulated?" "How well was this variable controlled by the experimenter?" "How well was this variable explained to participants?" "How well was this variable measured?"

"How well was this variable measured?"

Study details in question 43: Write a question you would ask of Dr. Phillips' study to interrogate each of the four validities.

Sample questions could include:a. Construct validity: "Was the math test given to the participants really a good measure of their understanding of multiplication?"b. External validity: "Were the participants chosen for the study like other second-graders, or were they different in some way?"c. Statistical validity: "Was there a difference between Group A and Group B?"d. Internal validity: "Is there some explanation for the difference in math scores other than instruction type?"

Which of the following is true of selection effects? Selection effects are a type of control variable. Selection effects can occur when experimenters allow participants to choose their own treatment group. Selection effects are unimportant for interrogating external validity. Selection effects are rarely considered in causal experiments

Selection effects can occur when experimenters allow participants to choose their own treatment group.

What type of experimental design is affected by order effects? What can researchers do to address order effects, and how does this fix them?

Within-groups designs are affected by order effects. Counterbalancing is the means by which researchers address order effects. Counterbalancing addresses order effects by varying the order in which the levels of the independent variable are presented to the participants. If one order is causing a different pattern of results, then the inclusion of another order should reveal that to be the case.

Considering a measure's face validity is a good way to interrogate the construct validity of the dependent variable. not necessary in experiments. only done if an experiment uses observational measures. the first step in establishing causation.

a good way to interrogate the construct validity of the dependent variable.

Of the following, which will produce the MOST narrow and precise confidence interval? a small sample size with less variability a small sample size with greater variability a large sample size with less variability a large sample size with more variability

a large sample size with less variability

Explain why experiments are better than your own personal experience at making causal claims.

Experiments provide a comparison group, whereas personal experience does not provide a comparison group. With experiments, the question "compared to what?" can be answered. One knows one's own experience in given situations, but there's no way of knowing if those experiences are different from or the same as other people's experiences in those same situations.

Explain the difference between full counterbalancing and partial counterbalancing. Why would a researcher choose partial counterbalancing over full counterbalancing?

Full counterbalancing occurs when all participants are exposed to all conditions/levels of the independent variable. Partial counterbalancing occurs when participants are exposed to only a subset of all the possible combinations. Partial counterbalancing should be chosen when there are four or more conditions/levels of the independent variable. There would be so many combinations of levels of the independent variable that it might not make sense to have participants exposed to all combinations.

Which of the following is a reason that researchers typically choose to prioritize internal over external validity? They want to be able to generalize results to the population. Having a confound-free setting allows them to make causal claims. Conducting an experiment in the laboratory is more expensive and time consuming. Researchers can seek out more diverse and representative participants.

Having a confound-free setting allows them to make causal claims.

An independent-groups design is also known as a design __________. between-subjects matched-groups within-groups mixed

between-subjects

Dr. Alfonse, a developmental psychologist, conducts a study to determine whether children prefer books with drawn illustrations or with photographs. A group of 30 preschoolers are shown two copies of a book (Ferdinand the Bull) at the same time. Although the story is the same, one book is illustrated with drawings and the other is illustrated with photos. Students are then asked to indicate which book they prefer. This is an example of which of the following designs? longitudinal design repeated-measures design concurrent-measures design posttest-only design

concurrent-measures design

A design in which 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 is known as a(n) staggered-administration design. concurrent-measures design. repeated-measures design. between-group design.

concurrent-measures design.

The question "Can the causal relationship generalize to other people, places, and times?" refers to what type of validity? internal face construct external

external

Random selection enhances ________ validity, and random assignment enhances ________ validity. internal; internal external; external internal; external external; internal

external; internal

Identify a major disadvantage to using matched groups when assigning participants. it removes randomness it requires more time and resources participants may know the person they are matched with the researcher will lose data from participants who don't match

it requires more time and resources

A variable that the researcher controls is a ________ variable. measured manipulated dependent selection

manipulated

Researchers conducting an experiment can ensure temporal precedence by doing which of the following? having a control group manipulating the cause before measuring the effect running a manipulation check establishing covariance

manipulating the cause before measuring the effect

Before she administers the independent variable to her two test groups, Dr. Mackintosh gives all participants an IQ test. She then pairs up each participant with another who had a similar IQ score. Based on the flip of a coin, one member of the pair is assigned to test Group A and one is assigned to test Group B. What technique is Dr. Mackintosh using? matched groups selection bias group assignment external validity

matched groups

Which of the following research designs is used to address possible selection effects? posttest-only designs matched-groups designs pretest/posttest designs correlational designs

matched-groups designs

When participants are sorted from lowest to highest on a variable, grouped into sets of two, and then one person from each set is assigned at random to each of the experimental groups, this is called random assignment. biased choice. matching. sorting.

matching.

In true experiments,________ is to dependent variable as ________ is to independent variable. measuring; manipulating controlling; manipulating recording; measuring manipulating; measuring

measuring; manipulating

Practice effects and carryover effects are examples of ________ effects. order scientific between-person causal

order

Which of the following is a threat to internal validity found in within-groups designs but not in independent-groups designs? selection effects design confounds demand characteristics practice effects

practice effects

A threat to internal validity occurs only if a potential design confound varies with the independent variable haphazardly. systematically. spontaneously. especially.

systematically.

What is the primary difference between pretest/posttest designs and within-groups designs? the number of participants used the number of times the dependent variable is measured the number of levels of the independent variable participants are exposed to there is no difference between the two designs.

the number of levels of the independent variable participants are exposed to

One reason researchers use within-group designs is to eliminate practice effects. it is less time-consuming for the participants. to increase the diversity of the participant pool. they require fewer participants.

they require fewer participants.

Which of the following phrases describes a manipulated variable? "Participants wrote down how many text messages they had sent the day before." "Researchers recorded whether participants volunteered to help the student in need or did not volunteer." "Participants were placed in the high tempo music condition, the low tempo music condition, or the no music condition based on which color card they randomly drew from a deck." "Researchers recorded the length of time participants took to complete the anagrams."

"Participants were placed in the high tempo music condition, the low tempo music condition, or the no music condition based on which color card they randomly drew from a deck."

Name three advantages of within-groups designs.

(a) Within-groups designs eradicate selection threats; (b) researchers get more precise estimates of the differences between groups because more of the extraneous error is removed; and (c) they typically require fewer participants.

Name three disadvantages of within-groups designs.

(a) Within-groups designs have the potential for order effects; (b) they have an increased risk for demand characteristics; (c) it may not be realistic to have participants provide repeated measures; and (d) it may be impossible for participants to experience something twice (as in the example from the textbook of learning to ride a bike twice).

Study details in question 43: Using Dr. Phillips' study as an example, explain the difference between a control group and a comparison group. Why did Dr. Phillips' study not have a control group?

A control group is a level of an independent variable that is intended to represent no treatment or a neutral condition. Dr. Phillips' study did not have a control group because a group without instruction would not have made sense in his study. A comparison group, by contrast, is a group that treatment groups can be compared with. In this case, Dr. Phillips can compare each of the four conditions with one another to determine which is the most effective way to teach multiplication.

What is a design confound? What type of validity is threatened by design confounds, and how is it threatened? Explain why not all problems in a study are necessarily confounds.

A design confound is a second variable that varies systematically with an independent variable. It threatens internal validity because it provides an alternate explanation for the results. In other words, students may say that although it appears that X causes Y, a design confound means that some other variable could actually be causing Y. Problems in a study are confounds only if they co-occur/vary systematically with the independent variable.

Define manipulation checks and pilot studies, and explain how they address the construct validity of an experiment.

A pilot study is a study conducted before the experimental study that tests whether the manipulation works (e.g., does happy music cause happiness?). Manipulation checks are conducted within the experimental study itself and ask the participants to report on their experiences (e.g., after listening to happy music, they are asked to state if they are happy before completing the dependent measure). Both manipulation checks and pilot studies address whether the independent variable was manipulated as intended. This ensures that the manipulation works. This is important to construct validity because it ensures that the proposed cause is in fact the cause.

What is the difference between a posttest-only design and a pretest/posttest design? Provide both a benefit and a disadvantage of using a pretest/posttest design.

A posttest-only design involves measuring the dependent variable once, whereas a pretest/posttest design involves measuring the dependent variable before and after the experimental manipulation. (Note that students should not state that the pretest/posttest design is a repeated-measures design or a within-groups design in their responses.)In each response, students must state one benefit of the pretest/posttest design, which could include: is good with small sample sizes; allows the researcher to ensure that random assignment worked; allows for the direct observation of change due to the manipulated variable. Students must also state one cost of the pretest/posttest design, which could include: can lead to suspicion among the participants (leading to demand characteristics); takes more time/adds an extra step; can change the behavior of the participants.

Study details in question 43: Explain why Dr. Phillips cannot conduct a within-groups design.

A within-groups design is not possible because being taught multiplication can only happen the first time once. As such, it would be impossible to have participants be taught for the first time four different times.

What is the difference between concurrent-measures designs and repeated-measures designs? Concurrent-measures designs are independent-groups designs; repeated-measures designs are within-group designs. Concurrent-measures designs expose participants to the levels of the independent variable at roughly the same time; repeated-measures designs expose participants to the levels of the independent variable sequentially. Concurrent-measures designs can have any number of levels of an independent variable; repeated-measures designs can only have two levels of the independent variable. There is no difference; they are two terms for the same design.

Concurrent-measures designs expose participants to the levels of the independent variable at roughly the same time; repeated-measures designs expose participants to the levels of the independent variable sequentially.

Which of the following is true of control variables? Control variables help define the control group. Control variables determine external validity. Control variables change the level of the manipulation. Control variables are kept the same for all participants.

Control variables are kept the same for all participants.

Explain why control variables are necessary in experiments and why they are not actually variables.

Control variables are not actually variables because they do not actually vary. Control variables are necessary in experiments to deal with threats to internal validity. Control variables are those aspects of a study that are held constant or equal across groups and that allow researchers to rule out alternate explanations for the causal relationships.

________ is used to control order effects in an experiment. Random assignment A design confound Increasing demand characteristics Counterbalancing

Counterbalancing

Dr. Kline, an environmental psychologist, conducts a study to examine whether visiting zoos causes people to have more positive attitudes toward environmental conservation. He asks a group of 45 people attending the zoo on a Saturday morning about their attitudes. He finds that 69% of the people report having a positive attitude after their visit. Which of the following is true? His control group is people who did not visit the zoo. He does not have a dependent variable. He can make a strong causal claim about the effect of zoo visits on environmental attitudes. His study does not qualify as an experiment.

His study does not qualify as an experiment.

Which of the following is true about replicating experiments? Every experiment should test something slightly different to get new information. The same group of people should participate every time. Only the original experimenters can do the replication. Multiple tests result in a better estimate of the population.

Multiple tests result in a better estimate of the population.

Study details in question 43: Name two ways that Dr. Phillips could avoid selection threat when he is assigning participants to conditions and explain why each of these methods avoids selection threat.

One method is random assignment. Example: rolling a die, picking a card. Random assignment avoids selection threat by spreading any individual differences randomly among the four conditions. The other method is matched-group designs. Example: IQ, mathematical ability, SES. Matched-groups designs avoid selection threat by making sure the groups are even on a particular variable. Note that students should not get credit for saying that a pretest/posttest design avoids selection threat. This design can detect a selection threat but can do nothing to avoid it.

Study details in question 43: Identify the independent variable(s), dependent variable(s), and a control variable in Dr. Phillips' study.

The independent variable is the learning style or type of instruction (students should be clear in their responses in stating that there is only one independent variable; it should not be confused with the four levels/conditions of the independent variable). The dependent variable is the number of multiplication problems solved correctly. Students may state several control variables, but they need to select only one. Options include keeping the number of test items the same, keeping the math problem difficulty level the same, and keeping the exposure to math instruction the same length (30 minutes).

RESEARCH STUDY 10.3: Dr. Phillips is a cognitive-developmental psychologist who studies how children learn mathematical principles. His current interest is in studying how children learn multiplication. He recruits 54 second-graders who have not yet learned how to multiply numbers to participate in his study. He is curious as to whether different teaching styles lead to better mastery of the subject. One condition (A) involves having students read a chapter on multiplication from their textbook. One condition (B) involves having students watch a video of cartoon characters explaining multiplication. One condition (C) involves having students being taught multiplication by a teacher. One condition (D) involves having students being taught multiplication by a fourth-grader. In all conditions, participants spend 30 minutes learning the fundamentals of multiplication and are then given the same 15-question multiplication test. The number of questions answered correctly is recorded. The data are below. Number of Multiplication Problems Solved Group A (Textbook) 4 Group B (Cartoon video) 10 Group C (Teacher) 12 Group D (Fourth-grader) 6 Groups B, C, D vs. Group A Statistically significant difference d = .27 Groups B, C vs. Group D Statistically significant difference d = .22 Group C vs. Group D Statistically significant difference d = .45 Group B vs. Group C No statistically significant difference d = .07 Group A vs. Group D No statistically significant difference d = .03 Based on the results of his study, Dr. Phillips hopes to make a causal statement. Explain how his study does or does not meet the three rules of causation.

The three rules for causation are covariance, temporal precedence, and internal validity. Dr. Phillips' study does meet the three rules of causation. In terms of covariance, there are differences between at least two of the groups, which indicates the manipulation had an effect on children's mathematical ability. In terms of temporal precedence, because Dr. Phillips manipulated the learning experience before measuring mathematical ability, he established temporal precedence. In terms of internal validity, Dr. Phillips removed other explanations for the link between learning experiences and mathematical ability, such as the length of the math test and the time exposed to the instruction.

Which of the following is an advantage of within-groups designs? These designs avoid order effects. These designs avoid demand characteristics. These designs are always possible. These designs rely on fewer participants.

These designs rely on fewer participants.

RESEARCH STUDY 10.2: Dr. Dormeur studies sleep and sleep disorders. She is curious as to whether technology exposure before bedtime causes people to fall asleep more slowly. She recruits a sample of 60 middle-aged women from a local church who reported no history of sleep problems. She creates three conditions. All participants come to the sleep lab for three nights in a row and experience all three conditions. In the first condition (A), participants were asked to play an online game (Candy Crush) on an iPad for 10 minutes prior to going to bed. In the second condition (B), participants were asked to read an article using an iPad that discussed tricks and tips for improving one's score on Candy Crush (which took about 10 minutes). In the third condition (C), participants were asked to read a newspaper article about the inventor of Candy Crush (which took about 10 minutes). With the use of an electroencephalograph (EEG), the researcher measures how long it takes participants to fall asleep. Dr. Dormeur was concerned that asking participants how long it took them to fall asleep would lead them to suspect that was the purpose of the study. Her decision to measure how long it took participants to go to sleep using the EEG instead of self-report was meant to decrease which of the following? order effects selection effects demand characteristics counterbalancing effects

demand characteristics

The d score is a standardized measure of the degree to which the independent variable caused a change in the dependent variable. This is also known as the impact factor. change rate. standard change. effect size.

effect size.

RESEARCH STUDY 10.2: Dr. Dormeur studies sleep and sleep disorders. She is curious as to whether technology exposure before bedtime causes people to fall asleep more slowly. She recruits a sample of 60 middle-aged women from a local church who reported no history of sleep problems. She creates three conditions. All participants come to the sleep lab for three nights in a row and experience all three conditions. In the first condition (A), participants were asked to play an online game (Candy Crush) on an iPad for 10 minutes prior to going to bed. In the second condition (B), participants were asked to read an article using an iPad that discussed tricks and tips for improving one's score on Candy Crush (which took about 10 minutes). In the third condition (C), participants were asked to read a newspaper article about the inventor of Candy Crush (which took about 10 minutes). With the use of an electroencephalograph (EEG), the researcher measures how long it takes participants to fall asleep. Dr. Dormeur's participants have agreed to participate in the study for only three nights each. Which type of counterbalancing can she use? partial counterbalancing full counterbalancing minimal counterbalancing carryover counterbalancing

full counterbalancing

Generally, what is the main priority for experimental studies? construct validity external validity internal validity statistical validity

internal validity

When a participant arrives for her experiment, Dr. Dajani flips a coin to determine which of two experimental conditions the participant will receive. Using this strategy increases the ________ of her study. external validity bias internal validity counterbalancing

internal validity

RESEARCH STUDY 10.2: Dr. Dormeur studies sleep and sleep disorders. She is curious as to whether technology exposure before bedtime causes people to fall asleep more slowly. She recruits a sample of 60 middle-aged women from a local church who reported no history of sleep problems. She creates three conditions. All participants come to the sleep lab for three nights in a row and experience all three conditions. In the first condition (A), participants were asked to play an online game (Candy Crush) on an iPad for 10 minutes prior to going to bed. In the second condition (B), participants were asked to read an article using an iPad that discussed tricks and tips for improving one's score on Candy Crush (which took about 10 minutes). In the third condition (C), participants were asked to read a newspaper article about the inventor of Candy Crush (which took about 10 minutes). With the use of an electroencephalograph (EEG), the researcher measures how long it takes participants to fall asleep. Which of the following designs is Dr. Dormeur using? pretest/posttest design concurrent-measures design repeated-measures design posttest-only design

repeated-measures design

Experiments use random assignment to avoid which of the following? random selection selection effects carryover effects demand characteristics

selection effects

RESEARCH STUDY 10.2: Dr. Dormeur studies sleep and sleep disorders. She is curious as to whether technology exposure before bedtime causes people to fall asleep more slowly. She recruits a sample of 60 middle-aged women from a local church who reported no history of sleep problems. She creates three conditions. All participants come to the sleep lab for three nights in a row and experience all three conditions. In the first condition (A), participants were asked to play an online game (Candy Crush) on an iPad for 10 minutes prior to going to bed. In the second condition (B), participants were asked to read an article using an iPad that discussed tricks and tips for improving one's score on Candy Crush (which took about 10 minutes). In the third condition (C), participants were asked to read a newspaper article about the inventor of Candy Crush (which took about 10 minutes). With the use of an electroencephalograph (EEG), the researcher measures how long it takes participants to fall asleep. Given that there are three conditions/levels of the independent variable, how many orders of the conditions are possible in Dr. Dormeur's study? three six nine twelve

six

Which of the following is a reason that a researcher might choose a pretest/posttest design? to ensure that random assignment made the treatment/comparison groups equal to determine between-group differences to allow for the study of spontaneous behaviors to improve construct validity

to ensure that random assignment made the treatment/comparison groups equal

Dr. West wants to know what kind of exercise makes people happiest. She randomly assigns participants to either a weight-lifting or cardio group. Each group does the assigned exercise for 30 minutes. They then report on their mood on a scale of 1-10. What could Dr. West do to eliminate potential confounds in her experiment? study only one type of exercise use the same room and exercise instructor for both groups let participants choose what type of exercise they want ahead of time have participants describe their emotions however they want instead of using a scale

use the same room and exercise instructor for both groups

Using a matched-group design is especially important in which of the following cases? when you have only a few people in your study when you have at least three levels/conditions of the independent variable when you do not have a control group when you have a complex dependent variable

when you have only a few people in your study