Week 3: Basic Design 2 - Simple Experiments That Have 1 Independent Variable

Question 1

How many levels do you need for a TRUE EXPERIMENT?

A: 1

B: 2

C: 3+

Answer 1

B: 2

We need a minimum of 2 levels. 3 or more is better because then we can determine a non-linear/functional relationship.

Question 2

Can you use either a quantitative or a qualitative independent variable?

A: Yes

B: No

Answer 2

A: Yes

EXAMPLE:
Research Question: Does our environment influence our well-being?

Hypothesis: A cluttered environment causes stress

Qualitative: Dirty coffee cups vs. Dirty food plates

Quantitative: Manipulate # of dirty dishes (How many levels? What range will we use?)

Question 3

What are the three different measures we can use for the dependent variable?

A: Self-report, behavioral, and physiological

B:

C:

Answer 3

A: Self-report, behavioral, and physiological

> Self-report (E.g., rate how stressed you feel)

> Behavioral (E.g., observe stress behaviors like fidgeting, nail-biting, etc.)

> Physiological (E.g., heart rate, cortisol)

Question 4

What are the four different scales of measurement we can use?

A:

B: Nominal, ordinal, interval, and ratio

C:

Answer 4

B: Nominal, ordinal, interval, and ratio

Nominal:
> Categorical
> E.g., categorize participants as “Stressed” or “Not Stressed”

Ordinal:
> Categorical
> E.g., not stressed, a little stressed, somewhat stressed, very stressed

Interval:
> Numerical
> E.g., rate level of stress on a scale of 1 (not stressed) to 50 (extremely stressed)

Ratio:
> Numerical
> E.g., count the number of stress behaviors

Question 5

What is test-retest reliability?

A:

B:

C: Test-retest reliability means that if they test a participant and then test them again, they want to make sure they are getting the same kind of response. In other words, if they test someone’s stress level with some kind of instrument, they should get a similar result every time they test that participant.

Answer 5

C: Test-retest reliability means that if they test a participant and then test them again, they want to make sure they are getting the same kind of response. In other words, if they test someone’s stress level with some kind of instrument, they should get a similar result every time they test that participant.

> Ensure any instruments/measures you use are dependable

> E.g., stress survey, instrument used to test participant heart rate

Question 6

What is interrater reliability?

A: Interrater reliability means that if they have multiple judges assessing behaviors, they want to make sure the judges come to similar conclusions about the participants. So if they have different people rating how stressed a participant seems based on their behaviors, they want the ratings to be consistent between the different judges.

B:

C:

Answer 6

A: Interrater reliability means that if they have multiple judges assessing behaviors, they want to make sure the judges come to similar conclusions about the participants. So if they have different people rating how stressed a participant seems based on their behaviors, they want the ratings to be consistent between the different judges.

> Ensure ratings are consistent, no matter which particular person is making them

> E.g., judges ratings of participant stress levels

Question 7

What is construct validity?

A:

B:

C: Construct validity asks whether the operational definitions chosen are appropriate. For example, if they define “clutter” operationally as having more than 10 dirty dishes in a room, construct validity would ask if having dirty dishes actually measures the construct of clutter. It also asks if their measure of stress (e.g. blood pressure) actually measures the construct of stress.

Answer 7

C: Construct validity asks whether the operational definitions chosen are appropriate. For example, if they define “clutter” operationally as having more than 10 dirty dishes in a room, construct validity would ask if having dirty dishes actually measures the construct of clutter. It also asks if their measure of stress (e.g. blood pressure) actually measures the construct of stress.

Clutter → dirty dishes appropriate?
Stress → ???

Question 8

What is external validity?

A:

B: External validity asks whether the study can be generalized beyond the materials and participants used in the study. It questions whether the results from a laboratory study, for example, can be generalized to the real world.

C:

Answer 8

B: External validity asks whether the study can be generalized beyond the materials and participants used in the study. It questions whether the results from a laboratory study, for example, can be generalized to the real world.

When defining variables and population, those choices will determine how widely you can generalize results.

Question 9

What is internal validity?

A: Internal validity asks whether, in a study with an independent variable and dependent variable if the differences in the dependent variable can actually be attributed to the manipulation of the independent variable. In other words, it asks if we can conclude that changes in the independent variable caused the changes in the dependent variable.

B:

C:

Answer 9

A: Internal validity asks whether, in a study with an independent variable and dependent variable if the differences in the dependent variable can actually be attributed to the manipulation of the independent variable. In other words, it asks if we can conclude that changes in the independent variable caused the changes in the dependent variable.

Does a cluttered environment (IV) cause stress (DV)

Question 10

What is an extraneous variable?

A:

B:

C: Extraneous variables are absolutely anything in the study that is not the independent variable and is not the dependent variable. Everything that is not the independent or dependent variable is considered an extraneous variable.

Answer 10

C: Extraneous variables are absolutely anything in the study that is not the independent variable and is not the dependent variable. Everything that is not the independent or dependent variable is considered an extraneous variable.

EXAMPLE:
> Rat running speed study
> The age of the rats, whether they were hungry, their gender, where they were from (e.g. Ohio vs Wyoming), and how they were handled.
> Anything not related to the independent variable (drug vs placebo) or dependent variable (running speed) could be considered an extraneous variable.

Question 11

What is a confounding variable?

A:

B:

C: A confounding variable is a type of extraneous variable that will cause havoc to the study’s internal validity. A confounding variable is an extraneous variable where the level of the variable systematically changes across levels of the independent variable. This is very bad for internal validity because it means the confounding variable is entangled with the independent variable, making it impossible to determine if outcomes are due to the independent variable or the confounding variable.

Answer 11

C: A confounding variable is a type of extraneous variable that will cause havoc to the study’s internal validity. A confounding variable is an extraneous variable where the level of the variable systematically changes across levels of the independent variable. This is very bad for internal validity because it means the confounding variable is entangled with the independent variable, making it impossible to determine if outcomes are due to the independent variable or the confounding variable.

EXAMPLE:
In the study looking at whether the color of text impacts reading speed, using different story topics (e.g. about a dog, cat, or cow) for each color condition would create a confounding variable. This is because the story topic would systematically vary depending on the color condition, making it impossible to tell if any differences were due to color or the topic.

Question 12

What are three ways to control an extraneous variable?

A: Keep it constant, vary it randomly, counterbalance it across conditions

B:

C:

Answer 12

A: Keep it constant, vary it randomly, counterbalance it across conditions

Question 13

What does “keep it constant” mean?

A:

B: One way to control an extraneous variable is to keep it constant, meaning no matter what condition a participant is assigned to, they will be exposed to the same level of the extraneous variable. As an example, in the rat running speed study, they could keep the age of the rats constant by only using rats that are 10 months old in both conditions.

C:

Answer 13

B: One way to control an extraneous variable is to keep it constant, meaning no matter what condition a participant is assigned to, they will be exposed to the same level of the extraneous variable. As an example, in the rat running speed study, they could keep the age of the rats constant by only using rats that are 10 months old in both conditions.

> The same value or level is maintained for all conditions

EXAMPLE:
> Violent TV study
> Manipulate the quantity of Violent TV (0, 5, 10 hours)
> Keep constant TYPE OF VIOLENT TV (E.g., only Tom and Jerry cartoons)

Question 14

What does “vary it randomly” mean?

A:

B: Another way to control an extraneous variable is to allow it to vary randomly. For example, in the rat study, they could randomly assign rats of varying ages between 2-10 months to the different conditions. This allows the extraneous variable to randomly vary across conditions.

C:

Answer 14

B: Another way to control an extraneous variable is to allow it to vary randomly. For example, in the rat study, they could randomly assign rats of varying ages between 2-10 months to the different conditions. This allows the extraneous variable to randomly vary across conditions.

> Level or value is randomly distributed across participants

> Most common way of dealing with an extraneous variable

> By randomly assigning participants to conditions we assume participant traits (e.g., talents, height, intelligence, hobbies) are more or less distributed across conditions.

Question 15

What does “counterbalance it across conditions” mean?

A: The third way to control an extraneous variable is to counterbalance it across conditions. This means making sure the extraneous variable, or its effects, are balanced or distributed equally across the different conditions. For example, in the rat study, they could counterbalance age by ensuring an equal number of young and old rats are assigned to each condition.

B:

C:

Answer 15

A: The third way to control an extraneous variable is to counterbalance it across conditions. This means making sure the extraneous variable, or its effects, are balanced or distributed equally across the different conditions. For example, in the rat study, they could counterbalance age by ensuring an equal number of young and old rats are assigned to each condition.

> COUNTER the effect of an EV by BALANCING it across
conditions

> E.g. Height of participants



Question 16

What does it mean when something varies systematically?

A:

B:

C: When an extraneous variable varies systematically across levels of the independent variable, it is very bad for internal validity. A systematic variation means the extraneous variable changes predictably depending on the level of the independent variable being tested. For example, if all the young rats were in one condition and all the old rats in another, their age would be varying systematically with the condition. This makes the extraneous variable difficult to disentangle from the independent variable.

Answer 16

C: When an extraneous variable varies systematically across levels of the independent variable, it is very bad for internal validity. A systematic variation means the extraneous variable changes predictably depending on the level of the independent variable being tested. For example, if all the young rats were in one condition and all the old rats in another, their age would be varying systematically with the condition. This makes the extraneous variable difficult to disentangle from the independent variable.

RAT STUDY SOLUTIONS:
> If you’re going to “vary randomly” you could still let age vary randomly but use many more participants
in each condition.

> If you’re going to “counterbalance” it’s more work-intensive than the other options BUT If we are really worried about age, counterbalancing is the best way to ensure equality across conditions. This would look like having a 2-month-old in both “no drug” and “drug” and so on with all the different ages between 2-10 months.

Question 17

What is a within-subjects design?

A:

B:

C: A within-subjects design means that each participant is exposed to all conditions of the study. So every participant experiences all levels of the independent variable, rather than different participants being assigned to different levels as in a between-subjects design.

Answer 17

C: A within-subjects design means that each participant is exposed to all conditions of the study. So every participant experiences all levels of the independent variable, rather than different participants being assigned to different levels as in a between-subjects design.

> All participants experience all levels of the Independent Variable

> The Stroop Experiment from your lab used this design

> E.g., all rats would take part in both levels of the Independent Variable (Placebo, Drug)

> Crossing over from one IV level to the other - everyone competes with themselves - the participant gets the placebo and the drug and their reaction is measured

PROS:
> Strongest test of the hypothesis
> Each participant serves as their own control
> Need fewer participants
> Less costly

CONS:
> Possible order effects
> More complex design for controlling extraneous
variables

Question 18

What is a between-subjects design?

A:

B: A between-subjects design means that each participant is only exposed to a single level of the independent variable. So different participants are assigned to each level or condition, rather than one participant experiencing all levels as in a within-subjects design.

C:

Answer 18

B: A between-subjects design means that each participant is only exposed to a single level of the independent variable. So different participants are assigned to each level or condition, rather than one participant experiencing all levels as in a within-subjects design.

> Participants experience only one level of the Independent Variable

> The Violent TV/Aggression study; kids watched either 0, 5, or 10 hours

> E.g., all rats would take part in only one level of the Independent Variable (Placebo, Drug)

> There is NO crossover from one IV level to the other - everyone is competing with other participants - the participant either gets only the placebo or only the drug and their reaction is measured to whichever level they got and compared to everyone else

PROS:
> No concern about order effects
> Easier to design

CONS:
> Less powerful test of hypothesis
> More participants
> More costly
> No guarantee that groups are truly equivalent



Question 19

What are order effects?

A: Order effects refer to when the order in which participants experience conditions is confounded with the levels of the independent variable. This can lead to problems like practice effects or fatigue effects carrying over between conditions. She emphasized that order needs to be controlled in within-subjects designs to avoid it becoming a confounding variable.

B:

C:

Answer 19

A: Order effects refer to when the order in which participants experience conditions is confounded with the levels of the independent variable. This can lead to problems like practice effects or fatigue effects carrying over between conditions. She emphasized that order needs to be controlled in within-subjects designs to avoid it becoming a confounding variable.

Question 20

What are the 3 types of order effects?

A: Practice effects, fatigue effects, and carryover effects

B:

C:

Answer 20

A: Practice effects, fatigue effects, and carryover effects

Question 21

What are practice effects?

A:

B:

C: Practice effects are an example of an order effect. In a within-subjects design where participants do one condition and then another, something they learn or get better at in the first condition due to practice could carry over and influence their performance in the second condition. This would be a practice effect as an unwanted order influence.

Answer 21

C: Practice effects are an example of an order effect. In a within-subjects design where participants do one condition and then another, something they learn or get better at in the first condition due to practice could carry over and influence their performance in the second condition. This would be a practice effect as an unwanted order influence.

Question 22

What are fatigue effects:

A:

B:

C: Fatigue effects are an example of an order effect. If participants do one condition and then another in a within-subjects design, their performance in the second condition could be impacted by fatigue from having already done the first condition. This would constitute a fatigue effect as an unwanted order influence.

Answer 22

C: Fatigue effects are an example of an order effect. If participants do one condition and then another in a within-subjects design, their performance in the second condition could be impacted by fatigue from having already done the first condition. This would constitute a fatigue effect as an unwanted order influence.

Question 23

What is a carryover effect?

A:

B: Practice effects and fatigue effects are examples of a broader category called carryover effects. Carryover effects refer to anything that might carry over from one condition to the next in a within-subjects design, influencing performance and threatening internal validity. This could include things they’ve learned, fatigue, or other unknown factors carrying over between conditions.

C:

Answer 23

B: Practice effects and fatigue effects are examples of a broader category called carryover effects. Carryover effects refer to anything that might carry over from one condition to the next in a within-subjects design, influencing performance and threatening internal validity. This could include things they’ve learned, fatigue, or other unknown factors carrying over between conditions.

Question 24

How do we prevent order effects in a within-subjects design?

A:

B:

C: To prevent order effects in a within-subjects design, the order in which participants experience the conditions needs to be counterbalanced. This means distributing the order of conditions evenly across participants, so that each condition is experienced first by some participants, second by others, and third by the remaining participants. This balances out any potential order influences across all conditions.

Answer 24

C: To prevent order effects in a within-subjects design, the order in which participants experience the conditions needs to be counterbalanced. This means distributing the order of conditions evenly across participants, so that each condition is experienced first by some participants, second by others, and third by the remaining participants. This balances out any potential order influences across all conditions.

• There is a visual of this on slide 40 that could be helpful

Question 25

True or false: If the specifics of the within-subjects study are such that order effects (practice, fatigue, or carryover effects) cannot be avoided then it should be conducted as a between-subjects design.

A: True

B: False

Answer 25

A: True

Question 26

How do we help to ensure we have equivalent groups in a between-subjects design?

A:

B:

C: We could have a large number of participants and randomly assign them to conditions. This helps distribute individual differences randomly. We could match participants on important individual differences (like birth order) and then randomly assign them to conditions. We could use stratified random sampling to ensure the proportions of subgroups (like gender) match the population in the sample.

Answer 26

C: We could have a large number of participants and randomly assign them to conditions. This helps distribute individual differences randomly. We could match participants on important individual differences (like birth order) and then randomly assign them to conditions. We could use stratified random sampling to ensure the proportions of subgroups (like gender) match the population in the sample.

NOTES:
> We usually solve this using random assignment
> Every participant has an equal chance of being in each condition
> Individual differences should be randomly distributed
> Most effective with a larger sample size

Question 27

What is a matched groups design?

A:

B: A matched groups design is when participants are matched or grouped based on some individual difference that is thought to be important, like birth order. The groups are then formed by matching participants on that characteristic, such as having equal numbers of firstborns, middle children, etc. The groups are then randomly assigned to conditions to help ensure equivalence between groups.

C:

Answer 27

B: A matched groups design is when participants are matched or grouped based on some individual difference that is thought to be important, like birth order. The groups are then formed by matching participants on that characteristic, such as having equal numbers of firstborns, middle children, etc. The groups are then randomly assigned to conditions to help ensure equivalence between groups.

MORE DETAILS:
A matched groups design would be used in a between-subjects study, where each participant is only exposed to one level of the independent variable. By matching and grouping participants before random assignment to conditions, it helps control for important individual differences in a between-subjects design.

EXAMPLE:
In the Violent TV study, you are worried about birth order so you….
1. Obtain birth order information
2. Randomly assign 1/3 of first borns to each condition; 1/3 of 2nd borns to each condition, etc.

Question 28

For the most part, what is the best way to control extraneous variables in a between-subjects design?

A:

B: The best way to control extraneous variables in a between-subjects design is to use a matched groups design. This involves matching participants on important individual differences thought to be extraneous variables, and then counterbalancing/distributing those variables across conditions through random assignment of the matched groups. This helps ensure the equivalence of groups and controls extraneous variables without the limitations of holding them constant.

C:

Answer 28

B: The best way to control extraneous variables in a between-subjects design is to use a matched groups design. This involves matching participants on important individual differences thought to be extraneous variables, and then counterbalancing/distributing those variables across conditions through random assignment of the matched groups. This helps ensure the equivalence of groups and controls extraneous variables without the limitations of holding them constant.

EXAMPLE:
> In the example from above for instance (Hypothesis: “People read green text more quickly than red text”), the best way to control the extraneous variable of topic/story content in a between-subjects design would be to counterbalance the topics across conditions.

> More specifically, having each topic (dog, cat, cow story) represented in each color condition to counterbalance any effects of the topic.

> This helps ensure the topic is not confounded with the color conditions being tested.

Question 29

For the most part, what is the best way to control extraneous variables in a within-subjects design?

A: The best way to control extraneous variables in a within-subjects design is through counterbalancing. This involves distributing the order of conditions evenly across participants so that each condition is experienced first by some, second by others, and third by the remaining participants. This balances out any potential influences of order or other extraneous variables across all conditions.

B:

C:

Answer 29

A: The best way to control extraneous variables in a within-subjects design is through counterbalancing. This involves distributing the order of conditions evenly across participants so that each condition is experienced first by some, second by others, and third by the remaining participants. This balances out any potential influences of order or other extraneous variables across all conditions.

EXAMPLE:
> For the reading speed study example in a within-subjects design, the best way to control the extraneous variable of topic/story content would be through counterbalancing.

> Specifically, having each topic (dog, cat, cow story) represented in each color condition for each participant cohort, and counterbalancing the order of conditions across cohorts.

> This ensures the topic is balanced across all color conditions and participants.

Question 30

What is the specific item effect:

A: A specific item effect occurs when the materials being used in a study are confounded or systematically varied with the levels of the independent variable. For example, using different story topics for each color condition in the reading speed study would create a specific item effect, since the topic would change depending on the color. This makes it impossible to determine if any differences are due to the color or the specific topic used.

B:

C:

Answer 30

A specific item effect occurs when the materials being used in a study are confounded or systematically varied with the levels of the independent variable. For example, using different story topics for each color condition in the reading speed study would create a specific item effect, since the topic would change depending on the color. This makes it impossible to determine if any differences are due to the color or the specific topic used.

Question 31

What are the 3 different types of bias in an experiment?

A:

B:

C: Experimenter, participant/subject, and materials

Answer 31

C: Experimenter, participant/subject, and materials

Question 32

What is experimenter bias?

A:

B: Experimenter bias refers to bias that comes from the researcher/experimenter conducting the study. It can be intentional, like deliberately influencing participants’ responses, which is considered a serious ethical violation. But it can also be unintentional, like subtly cueing participants through body language or tone of voice about which condition is expected to perform better.

C:

Answer 32

B: Experimenter bias refers to bias that comes from the researcher/experimenter conducting the study. It can be intentional, like deliberately influencing participants’ responses, which is considered a serious ethical violation. But it can also be unintentional, like subtly cueing participants through body language or tone of voice about which condition is expected to perform better.

POTENTIAL SOLUTIONS:
To prevent experimenter bias, it’s important for the researcher interacting with participants to be “blind” to the conditions. This means they don’t know which participants are in which condition when administering the study. One way to achieve this is through double-blinding, where neither the participants nor the experimenter know which condition a participant is in until after data collection and analysis is complete.

Question 33

What is participant/subject bias?

A:

B:

C: Participant/subject bias refers to when participants know the hypothesis of the study and may try to help the researcher get the desired results. This could influence their responses or behaviors in a way that biases the outcomes.

Answer 33

C: Participant/subject bias refers to when participants know the hypothesis of the study and may try to help the researcher get the desired results. This could influence their responses or behaviors in a way that biases the outcomes.

POTENTIAL SOLUTION:
Participants should be “blind” to the hypothesis and not know which condition they are in (placebo vs drug, etc.) to avoid any potential influence on the results.

Question 34

What is self-selection bias, a specific type of subject bias?

A: Self-selection bias occurs when participants are allowed to choose which condition they want to be in.

B:

C:

Answer 34

A: Self-selection bias occurs when participants are allowed to choose which condition they want to be in.

EXAMPLE:
A study comparing warm and cold water temperatures, where allowing participants to choose could result in fundamentally different types of people selecting each condition.

POTENTIAL SOLUTION:
Researchers should randomly assign participants to conditions rather than letting them self-select.

> You could ask people who’s willing to get into a 30-degree pool and then split them in half, half go into the 30-degree pool and the other half go into the 80-degree pool

Question 35

What is materials bias?

A: Biases that are inherent in the materials or measures used in a study. They stem from issues or influences within the content or format of the materials themselves. The materials are getting in the way of assessing the impact of the IV on the DV
B:

C:

Answer 35

A: Biases that are inherent in the materials or measures used in a study. They stem from issues or influences within the content or format of the materials themselves. The materials are getting in the way of assessing the impact of the IV on the DV



Question 36

What are the 4 different types of materials bias?

A:

B: Ceiling effect, floor effect, demand characteristic, and specific item effect.

C:

Answer 36

B: Ceiling effect, floor effect, demand characteristic, and specific item effect.

Question 37

Regarding materials bias, what is the ceiling effect?

A:

B:

C: A ceiling effect occurs when the materials or measures used in a study are too easy so that all participants in all conditions are performing very well or at the maximum level. This makes it difficult to detect any differences between conditions since everyone is clustered at the top or “ceiling” of performance.

Answer 37

C: A ceiling effect occurs when the materials or measures used in a study are too easy so that all participants in all conditions are performing very well or at the maximum level. This makes it difficult to detect any differences between conditions since everyone is clustered at the top or “ceiling” of performance.

EXAMPLE:
> Hypothesis: “Sleep affects math performance.”
> IV: Sleep (0 hours, 4 hours, 8 hours)
> DV: % Correct, Math Test
> Provided Addition Problems (E.g., 5+3)
> The participants did 2-digit addition problems after different amounts of sleep. If all groups were getting close to 100% correct on these problems, the materials would be too easy and cause a ceiling effect, making it hard to see if sleep made a difference since everyone scored highly.

POTENTIAL SOLUTION:
One way to address a ceiling effect is to pilot test materials on a sample group first, before using them in the full study. This allows researchers to check if the materials are too easy or difficult. If a ceiling effect is found, where all participants perform very well, the materials would need to be made more challenging in order to better detect potential differences between conditions.

Question 38

Regarding materials bias, what is the floor effect?

A:

B:

C: A floor effect occurs when the materials or measures used in a study are too difficult so that all or most participants in all conditions are performing very poorly or at the minimum level. This makes it hard to detect differences between conditions since everyone is clustered at the bottom or “floor” of performance.

Answer 38

C: A floor effect occurs when the materials or measures used in a study are too difficult so that all or most participants in all conditions are performing very poorly or at the minimum level. This makes it hard to detect differences between conditions since everyone is clustered at the bottom or “floor” of performance.

EXAMPLE:
> Hypothesis: “Sleep affects math performance.”
> IV: Sleep (0 hours, 4 hours, 8 hours)
> DV: % Correct, Math Test
> Provided calculus problems
> If the problems were so hard that nobody was getting any correct, it could cause a floor effect where differences between conditions can’t be detected because performance is too low across the board.

POTENTIAL SOLUTION:
One solution is to pilot test materials on a sample group first. This allows researchers to check if the materials are causing a floor effect where most or all participants perform poorly. If identified, the materials would need to be made less challenging in order to better detect potential differences between conditions in the full study.

Question 39

Regarding materials bias, what is the demand characteristic?

A:

B:

C:

Answer 39

C: Demand characteristics refer to cues or influences in the wording, format, or administration of materials that could lead participants to figure out the researcher’s hypothesis and respond in a way that supports what they think the researcher wants to find. This can bias responses and threaten the internal validity of a study.

EXAMPLE:
> A good example is survey questions.
> A study where one question asked about dating frequency, while another rated happiness. But the order of these questions was varied - and people reported lower happiness when asked about dating first, suggesting the dating question primed them to think more negatively and rate their happiness lower due to demand characteristics.
> Or, what if you were asked if you would help an elderly person who fell. Even if the real answer is no, it’s likely that you would say yes

POTENTIAL SOLUTION:
> Pilot testing materials to check for unintended cues or influences.
> Blinding participants to the true purpose and hypotheses of a study to prevent them from consciously or unconsciously responding in a way they think supports the researcher’s expectations.
> Counterbalancing question order to control for demand characteristics.

Question 40

Regarding materials bias, what is the specific item effect?

A: A specific item effect occurs when the materials or stimuli used in a study are confounded or systematically varied with the levels of the independent variable.

B:

C:

Answer 40

A: A specific item effect occurs when the materials or stimuli used in a study are confounded or systematically varied with the levels of the independent variable.

EXAMPLE:
The reading speed study compared reading green, red, and black text. Using different story topics (dog, cat, cow) for each color condition would create a specific item effect since the story topic was confounded with and varied systematically based on the text color condition. This made it unclear if any differences were due to the color or the specific story topic used.

POTENTIAL SOLUTION:
Counterbalancing the materials or stimuli across conditions. For example, in the reading speed study, having each story topic (dog, cat, cow) represented in each color condition. This ensures the topics are balanced and not predictive of the independent variable, avoiding specific item effects confounding the results.



Question 41

What is the importance of sampling from the population?

A:

B: When conducting research, researchers often want to generalize their findings beyond just the specific participants in their study to make inferences about a broader target population. To do this, researchers need to sample from the population in a way that allows the sample to represent the characteristics of the overall population.

C:

Answer 41

B: When conducting research, researchers often want to generalize their findings beyond just the specific participants in their study to make inferences about a broader target population. To do this, researchers need to sample from the population in a way that allows the sample to represent the characteristics of the overall population.

MORE DETAILS:
> What is the population you’re interested in?
> All the people you want to generalize to
> If the population is small enough, you can test everyone (E.g., Students enrolled in 100B this quarter)
> But usually you want to generalize more widely
> So you have to take a sample!
> The sample should represent the characteristics of the
population.

Question 42

What are 4 different types of sampling methods researchers can use?

A: Simple random sampling, convenience sampling, cluster sampling, and stratified proportional random sampling.

B:

C:

Answer 42

A: Simple random sampling, convenience sampling, cluster sampling, and stratified proportional random sampling.

Question 43

What is simple random sampling?

A:

B: Simple random sampling is a method where every individual in the target population has an equal probability of being selected for the sample. The key aspect is that it is truly random, with no bias, so that each person has the same chance of selection. However, it can be difficult to implement a truly simple random sample for large populations where not every person can feasibly be accessed.

C:

Answer 43

B: Simple random sampling is a method where every individual in the target population has an equal probability of being selected for the sample. The key aspect is that it is truly random, with no bias, so that each person has the same chance of selection. However, it can be difficult to implement a truly simple random sample for large populations where not every person can feasibly be accessed.

NOTE:
For very large populations like all adults in the United States, it is often not possible to truly do a simple random sample, since there is no way to randomly access every single individual with an equal chance of selection.

Question 44

What is convenience sampling?

A:

B: Convenience sampling is a type of non-probability sampling where participants are selected because they are convenient to the researcher. For instance, using participants who are easily contactable or accessible, such as students in a class. While convenient, it is not the most rigorous type of sampling since the participants may not accurately represent the target population.

C:

Answer 44

B: Convenience sampling is a type of non-probability sampling where participants are selected because they are convenient to the researcher. For instance, using participants who are easily contactable or accessible, such as students in a class. While convenient, it is not the most rigorous type of sampling since the participants may not accurately represent the target population.

MORE DETAILS:
> Participants are volunteers willing to come to the study
> Easily accessible
> E.g., Fliers in psychology building restrooms
> E.g., Participation for credit in psychology courses
> E.g., Amazon’s Mechanical Turk for online volunteers
> Not as representative as Simple Random Sampling but often it’s the best we can do
> Limits external validity

Question 45

What is cluster sampling?

A: Cluster sampling involves dividing the population into distinct groups or “clusters” and then randomly selecting some of those clusters for inclusion in the sample. Within each selected cluster, all individuals or elements are then included in the sample. This allows for ensuring the representation of subgroups within the overall population. For instance, sampling geographic regions to represent views across different areas.

B:

C:

Answer 45

A: Cluster sampling involves dividing the population into distinct groups or “clusters” and then randomly selecting some of those clusters for inclusion in the sample. Within each selected cluster, all individuals or elements are then included in the sample. This allows for ensuring the representation of subgroups within the overall population. For instance, sampling geographic regions to represent views across different areas.

MORE DETAILS:
> Ensuring clusters within the population are represented in the sample
> E.g., Surveying viewers of Netflix; may want to get an equal number of participants from North, South, East, and West areas of the USA.
> These four geographic areas are considered clusters
> Randomly sample from each cluster

Question 46

What is stratified proportional random sampling?

A:

B:

C: Stratified random sampling is a method where the population is divided into subgroups or “strata” based on attributes that are important to the study. Then a random sample is selected from each stratum in a number proportional to the stratum’s size in the population. This ensures that sample proportions match population proportions on key attributes. For instance, sampling language groups to maintain representation of monolingual, bilingual, and multilingual populations.

Answer 46

C: Stratified random sampling is a method where the population is divided into subgroups or “strata” based on attributes that are important to the study. Then a random sample is selected from each stratum in a number proportional to the stratum’s size in the population. This ensures that sample proportions match population proportions on key attributes. For instance, sampling language groups to maintain representation of monolingual, bilingual, and multilingual populations.

MORE DETAILS:
> Participants are randomly selected from specific
demographic categories (or strata)
> Proportionate to their frequencies in the population
> Let’s say n = 1000
> If our USA population of interest is 75% Monolinguals we would have 750 Monolingual participants
> If our USA population of interest is 20% Bilinguals we would have 200 Bilingual participants
> If our USA population of interest is 05% Multilinguals we would have 50 Multilingual participants

Question 47

CAREFUL… What’s the difference between “random sampling” and “random assignment?”

A: “Random sampling” refers to selecting a sample of participants from a larger population in a way that gives all individuals an equal chance of being selected. This helps ensure the sample is representative of the population. “Random assignment” refers to allocating participants who have already been selected or volunteered for a study into different conditions or groups in a random way. This helps ensure any individual differences between participants are distributed evenly across conditions by chance.

B:

C:

Answer 47

A: “Random sampling” refers to selecting a sample of participants from a larger population in a way that gives all individuals an equal chance of being selected. This helps ensure the sample is representative of the population. “Random assignment” refers to allocating participants who have already been selected or volunteered for a study into different conditions or groups in a random way. This helps ensure any individual differences between participants are distributed evenly across conditions by chance.