Week 2: Basic Design (True Experiment) Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

What is a true experiment?

A: In a true experiment we manipulate one variable and measure its effect on the other variable. The variable that is manipulated is called the independent variable, and the variable that is measured is called the dependent variable. We manipulate the independent variable to assess its effect on the dependent variable and everything else must be controlled.

B: A true experiment involves manipulating both variables simultaneously and observing the combined effects on each other without distinguishing between an independent or dependent variable.

C: In a true experiment, variables are left uncontrolled, allowing for natural fluctuations to influence the relationship between the manipulated and measured variables.

A

A: In a true experiment we manipulate one variable and measure its effect on the other variable. The variable that is manipulated is called the independent variable, and the variable that is measured is called the dependent variable. We manipulate the independent variable to assess its effect on the dependent variable and everything else must be controlled.

  1. Independent Variable (IV):
    > Manipulated variable
  2. Dependent Variable (DV):
    > Measured variable

EXAMPLE:
> Hypothesis: “Watching violent television CAUSES aggressive behavior.”
> Independent variable = violent TV (this is what we manipulated, so maybe one group watches 0 hours of violent TV and the other watches 10 hours per week and then you’ll compare these two groups)
> Dependent variable = aggression (this is what we’re measuring)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the characteristics of an independent variable?

A: An independent variable is a constant factor that remains unchanged throughout an experiment, providing a stable background against which the dependent variable can be measured.

B: The independent variable is solely determined by the participants themselves, without any manipulation by the researcher, allowing for natural variations in the variable without control or random assignment.

C: The independent variable is manipulated by the researcher, meaning they create conditions and assign participants to experience different levels of the variable. It must have a minimum of two levels or conditions. Participants must be randomly assigned to the levels of the independent variable. The independent variable must be operationally defined in detail.

A

C: The independent variable is manipulated by the researcher, meaning they create conditions and assign participants to experience different levels of the variable. It must have a minimum of two levels or conditions. Participants must be randomly assigned to the levels of the independent variable. The independent variable must be operationally defined in detail.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What does the term manipulation mean?

A: Manipulation refers to participants having complete control over the experimental conditions, deciding on the levels they want to experience without any intervention from the researcher.

B: The researcher is going to create conditions that participants will be assigned to experience and two levels are the absolute minimum (0 minutes of violent TV vs. 10 minutes of violent TV).

C: The term manipulation is misunderstood as referring to the fabrication of data, allowing researchers to manipulate or alter the results to fit their expectations without regard for the actual conditions experienced by participants.

A

B: The researcher is going to create conditions that participants will be assigned to experience and two levels are the absolute minimum (0 minutes of violent TV vs. 10 minutes of violent TV).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does the term random assignment mean?

A: Random assignment involves letting participants choose the experimental conditions they want to experience, ensuring that their personal preferences determine the levels of independent variables.

B: Participants are deliberately placed in conditions based on certain characteristics or traits that the researcher believes are relevant to the study.

C: Participants must be randomly assigned to the levels of our independent variables. This is a fundamental tenet of experimental design. There should be nothing about the participants themselves that determines what condition they will experience. So our participants will be randomly assigned to the conditions or levels of our independent variables (0 minutes of violent TV vs. 10 minutes of violent TV).

A

C: Participants must be randomly assigned to the levels of our independent variables. This is a fundamental tenet of experimental design. There should be nothing about the participants themselves that determines what condition they will experience. So our participants will be randomly assigned to the conditions or levels of our independent variables (0 minutes of violent TV vs. 10 minutes of violent TV).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is a construct?

A: A construct is a specific and precisely defined variable that leaves no room for interpretation or variation. It represents a concrete and universally agreed-upon concept in research.

B: A general concept or idea that is broad and not concretely defined. Constructs can mean different things to different people. Examples: intelligence, motivation, anxiety, and safety.

C: A construct is thought to be a highly specific and narrowly defined idea that is universally agreed upon, lacking the inherent subjectivity and variability associated with broad concepts like intelligence or motivation.

A

B: A general concept or idea that is broad and not concretely defined. Constructs can mean different things to different people. Examples: intelligence, motivation, anxiety, and safety.

EXAMPLE:
Independent variable = Violent TV
Dependent variable = Aggression

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is an operational definition?

A: An operational definition is a vague and imprecise description of a variable, allowing for subjective interpretation and lacking the need for concrete measurement or manipulation.

B: An operational definition takes a construct and gives much more detail about how that variable will be studied and measured. An operational definition provides a concrete, quantitative, or qualitative way to manipulate an independent variable or measure a dependent variable. It allows others to replicate a study based on the precise description of how variables are defined and measured.

C: The operational definition is perceived as a rigid and inflexible set of rules dictating how a variable should be studied and measured, leaving no room for adaptability or variations in research approaches.

A

B: An operational definition takes a construct and gives much more detail about how that variable will be studied and measured. An operational definition provides a concrete, quantitative, or qualitative way to manipulate an independent variable or measure a dependent variable. It allows others to replicate a study based on the precise description of how variables are defined and measured.

EXAMPLE:
Independent variable:
> Quantitative (# of hours of violent TV watched per week, 0 vs. 10 for example)
> Qualitative (type of violent TV watched, cartoons, news, movie for example - everyone would watch the same amount but the type of violence they are watching will be different).

Dependent variable:
> # of times a child hits another child on the playground
> # of children in their class that describe them as mean
> > Ask teachers to rate their behavior on a certain scale
> People assess their own thoughts, feelings, beliefs,
habits, skills, etc.
> Observe participants and measure their behavior
> Biological measurements (heart rate, blood pressure, cortisol levels)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is a quantitative manipulation?

A: A quantitative manipulation looks at how much of something a participant will receive or experience. In the example of violent TV exposure, it would involve assigning different groups to watch different amounts of violent TV, like zero hours per week versus 10 hours per week. A quantitative manipulation measures different amounts or levels of the independent variable.

B: A quantitative manipulation involves manipulating variables based solely on qualitative differences, disregarding any consideration for the quantity or amount of the variable experienced by participants.

C: A quantitative manipulation is thought to involve manipulating variables in a way that doesn’t measure different amounts or levels, neglecting the essential aspect of quantifying the independent variable.

A

A: A quantitative manipulation looks at how much of something a participant will receive or experience. In the example of violent TV exposure, it would involve assigning different groups to watch different amounts of violent TV, like zero hours per week versus 10 hours per week. A quantitative manipulation measures different amounts or levels of the independent variable.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What type of graph would be appropriate for a quantitative manipulation?

A: Line graph

B: Bar graph

C: Histogram

A

A: Line graph

> A line graph is appropriate for data from a quantitative manipulation because the midpoints along the x-axis are meaningful. When groups experience different amounts of the independent variable, like 0 hours and 10 hours of violent TV, there is a meaningful connection between those points on a continuum. A line can be drawn connecting the data points, and predictions can be made for intermediate values like 5 hours.

> There is a meaningful halfway point between 0 hours and 10 hours - it’s 5 hours

> Only quantitative manipulations can show functional relationships, not qualitative manipulation

> The independent variable will always be on the x-axis (horizontal)

> The dependent variable will always be on the y-axis (vertical)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a qualitative manipulation?

A: Manipulating variables solely based on the amount or quantity, ignoring any distinctions in the types or categories of the variable experienced by participants.

B: A qualitative manipulation is perceived as a method that doesn’t involve comparing different categories or kinds of the independent variable, overlooking the critical aspect of qualitative differences.

C: A qualitative manipulation differentiates groups based on the type of thing they experience, rather than the amount. In the example of violent TV, it could involve comparing watching violent cartoons versus TV crime shows. A qualitative manipulation compares different categories or kinds of the independent variable, rather than different amounts.

A

C: A qualitative manipulation differentiates groups based on the type of thing they experience, rather than the amount. In the example of violent TV, it could involve comparing watching violent cartoons versus TV crime shows. A qualitative manipulation compares different categories or kinds of the independent variable, rather than different amounts.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

True or false: Only using two conditions is limiting?

A: True

B: False

A

A: True

> Only using two conditions for a quantitative manipulation is somewhat limiting, because it only provides two data points. With only two points, the relationship can only be a straight line - either increasing, decreasing, or staying the same. Using three or more conditions allows the researcher to check if there is a nonlinear or functional relationship between the variables, rather than just an assumption of a straight linear relationship. It provides more data points to reveal potential nuances in the relationship.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is a nonlinear/functional relationship?

A: A nonlinear/functional relationship always represents a perfectly straight line, with no variation or curvature in the relationship between two variables.

B: A nonlinear or functional relationship refers to when the relationship between two variables is not a straight linear line. Using three or more levels of the independent variable allows the researcher to check if the data shows a nonlinear pattern, such as aggression increasing dramatically from 0 to 5 hours of TV exposure but leveling off from 5 to 10 hours, rather than steadily increasing the whole time. It reveals if the relationship has a more complex, nonlinear shape rather than simply going up or down in a straight line.

C: A nonlinear/functional relationship is thought to be limited to situations where the relationship is chaotic and unpredictable, disregarding the possibility of a structured and complex pattern.

A

B: A nonlinear or functional relationship refers to when the relationship between two variables is not a straight linear line. Using three or more levels of the independent variable allows the researcher to check if the data shows a nonlinear pattern, such as aggression increasing dramatically from 0 to 5 hours of TV exposure but leveling off from 5 to 10 hours, rather than steadily increasing the whole time. It reveals if the relationship has a more complex, nonlinear shape rather than simply going up or down in a straight line.

> These types of relationships are easier to see with 3 or more conditions. Actually, 3+ levels is the only way we can see a nonlinear/functional relationship.

> For example: 0 hours/week, 5 hours/week, and 10 hours/week

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Why does the range (minutes vs. hours) we choose to study matter?

A: The range is irrelevant, assuming that any range, whether narrow or wide, would provide an equally comprehensive understanding of the relationship between variables.

B: The range we choose to study matters because if the range is too narrow, it may not reveal the true nature of the relationship between the variables. As an example, if the range studied was 0 minutes to 10 minutes of TV exposure, there may be no observable difference, but that narrow range doesn’t show the full picture of what could happen over a wider range like 0 to 10 hours. A narrow range provides a limited view like a close-up photo, while a wider range gives more insight into the overall relationship.

C: The range selected is thought to have no impact on the ability to observe differences between variables, negating the idea that a wider range could offer more comprehensive insights than a narrow one.

A

B: The range we choose to study matters because if the range is too narrow, it may not reveal the true nature of the relationship between the variables. As an example, if the range studied was 0 minutes to 10 minutes of TV exposure, there may be no observable difference, but that narrow range doesn’t show the full picture of what could happen over a wider range like 0 to 10 hours. A narrow range provides a limited view like a close-up photo, while a wider range gives more insight into the overall relationship.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

So with quantitative manipulation, what are your two biggest concerns?

A: In quantitative manipulation, the number of levels or conditions of the independent variable is deemed irrelevant, and any number of levels, even just two, is considered sufficient for a thorough examination of the relationship between variables.

B: The range of data chosen in quantitative manipulation has no impact on the validity of the study, and a narrow range is thought to be just as effective in revealing the true nature of the relationship as a wider range.

C: The number of levels or conditions of the independent variable. At least three levels are needed to check for nonlinear relationships. And the range of data chosen. The range needs to be wide enough to potentially reveal the true nature of the relationship between variables, rather than a narrow range that may show no observable effects.

A

C: The number of levels or conditions of the independent variable. At least three levels are needed to check for nonlinear relationships. And the range of data chosen. The range needs to be wide enough to potentially reveal the true nature of the relationship between variables, rather than a narrow range that may show no observable effects.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What type of graph would be appropriate for a qualitative manipulation?

A: Line graph

B: Bar graph

C: Histogram

A

B: Bar graph

> A bar graph is appropriate for data from a qualitative manipulation because there is no meaningful connection or continuum between the categories. With a qualitative manipulation comparing different types rather than amounts, like violent cartoons vs. crime shows, there is no halfway point between the categories like there is with quantitative amounts. Since there is no continuum across discrete categories, a bar graph is more suitable than a line graph to display the data.

> There is no meaningful halfway point between cartoons and the news

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are the characteristics of a dependent variable?

A: It is the factor that is measured by the researcher to assess changes or effects. It needs to be operationally defined precisely to guide how it will be measured in the study. The scale of measurement used to collect data on the dependent variable needs to be identified as nominal, ordinal, interval, or ratio.

B: The dependent variable is thought to be a factor measured by participants themselves, without the need for operational definitions or precise guidance from the researcher.

C: The scale of measurement for the dependent variable is perceived as irrelevant to the study of the independent variable, disregarding the importance of identifying the measurement scale for accurate data collection.

A

A: It is the factor that is measured by the researcher to assess changes or effects. It needs to be operationally defined precisely to guide how it will be measured in the study. The scale of measurement used to collect data on the dependent variable needs to be identified as nominal, ordinal, interval, or ratio.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What must your dependent variable operational definition include?

A: The dependent variable operational definition is unnecessary, assuming that the type of measure and scale of measurement have no bearing on accurately assessing changes or effects.

B: The operational definition of the dependent variable is thought to only require a general description, without specifying the type of measure or the scale of measurement used.

C: The type of measure that will be used (self-report, behavioral, physiological). The specific scale of measurement that will be employed (nominal, ordinal, interval, ratio) including details of the scale used.

A

C: The type of measure that will be used (self-report, behavioral, physiological). The specific scale of measurement that will be employed (nominal, ordinal, interval, ratio) including details of the scale used.

17
Q

What are the three types of measurement?

A: Self-report, behavioral, and physiological.

B: Nominal, ordinal, and interval

C: The three types of measurement are thought to encompass only qualitative measures

A

A: Self-report, behavioral, and physiological.

> You need to determine the type of measurement you will be using in your study before determining the scale of measurement you will use.

> This is part of the operational definition for your dependent variable

18
Q

Describe the self-report measurement:

A: A method where researchers assess the thoughts, feelings, beliefs, skills, or abilities of participants without involving the participants’ own evaluations or ratings.

B: When the participant assesses their own thoughts, feelings, beliefs, skills, or abilities. An example would be asking someone to rate their own memory or how fast they think they can run a mile. Problems with self-report include that people may miscalculate their own behavior or attributes, or may not want to reveal things truthfully. We try not to rely too much on self-report and focus more on behavioral instead.

C: Self-report measurement is thought to solely rely on objective observations and behavioral assessments, excluding the participants’ own subjective evaluations of their internal states or attributes.

A

B: When the participant assesses their own thoughts, feelings, beliefs, skills, or abilities. An example would be asking someone to rate their own memory or how fast they think they can run a mile. Problems with self-report include that people may miscalculate their own behavior or attributes, or may not want to reveal things truthfully. We try not to rely too much on self-report and focus more on behavioral instead.

19
Q

Describe the behavioral measurement:

A: A method where researchers solely rely on participants’ self-reports and subjective evaluations of their behavior, excluding direct observations or measurements.

B: When the researcher directly observes and measures participants’ behavior. This avoids relying on self-reporting. Examples: timing how fast someone can run a mile, giving participants a word list to learn and then testing recall, or counting observed instances of impulse shopping behavior rather than asking about it.

C: Behavioral measurement is thought to involve asking participants about their behavior through surveys or interviews, neglecting the need for direct observation and measurement by the researcher.

A

B: When the researcher directly observes and measures participants’ behavior. This avoids relying on self-reporting. Examples: timing how fast someone can run a mile, giving participants a word list to learn and then testing recall, or counting observed instances of impulse shopping behavior rather than asking about it.

20
Q

Describe the physiological measurement:

A: A method that solely relies on self-report or behavioral observation, excluding the use of biological measures to assess participants’ physiological responses.

B: Physiological measurement is thought to involve subjective assessments of participants’ physiological states, neglecting the need for direct and objective measures such as heart rate, pupil dilation, blood pressure, or cortisol levels.

C: A biological measure of participants. Examples: measuring heart rate, pupil dilation, blood pressure, or cortisol levels. These provide objective data not based on self-report or behavioral observation alone.

A

C: A biological measure of participants. Examples: measuring heart rate, pupil dilation, blood pressure, or cortisol levels. These provide objective data not based on self-report or behavioral observation alone.

21
Q

What are the four scales of measurement?

A: Nominal, ordinal, interval, and ratio.

B: Basic, intermediate, advanced, and complex

C: Categorical, ordinal, continuous, and proportional

A

A: Nominal, ordinal, interval, and ratio.

> Once you’ve determined the type of measurement you’ll be using for your study you can determine the scale of measurement you’ll be using

> This is part of the operational definition for your dependent variable

22
Q

Describe the nominal scale of measurement:

A: The nominal scale of measurement is the simplest form, involving qualitative categories where measurements are either in or out of the category (all or nothing). There is no order or continuum between categories. Examples: yes/no questions about owning a cat or which month a person was born in. No continuum exists.

B: A method that involves ranking categories in a specific order, contrary to the correct description of having no order or continuum between categories.

C: The nominal scale of measurement is thought to encompass only quantitative measurements, neglecting the characteristic of involving qualitative categories with no order or continuum.

A

A: The nominal scale of measurement is the simplest form, involving qualitative categories where measurements are either in or out of the category. There is no order or continuum between categories. Examples: yes/no questions about owning a cat or which month a person was born in.

23
Q

Describe the ordinal scale of measurement:

A: A method that involves assigning numerical values to categories, contrary to the correct description of providing orders or rankings without defined numerical differences.

B: The ordinal scale of measurement is thought to encompass only qualitative measurements, neglecting the characteristic of providing a continuum with orders or rankings.

C: The ordinal scale provides categories with orders or rankings, but still no defined differences between categories. Example: rating mint ice cream preferences on a scale from “never” to “a lot”, or ranking runners who finished a race in first, second, or third place order, but not providing information on differences in times between placements. A continuum exists.

A

C: The ordinal scale provides categories with orders or rankings, but still no defined differences between categories. Example: rating mint ice cream preferences on a scale from “never” to “a lot”, or ranking runners who finished a race in first, second, or third place order, but not providing information on differences in times between placements. A continuum exists.

24
Q

Describe the interval scale of measurement:

A: The interval scale of measurement relies on qualitative data, assigning arbitrary numbers without assuming equal differences between them.

B: In the interval scale, the numerical values have a true zero, and differences between numbers represent proportional changes in the measured variable.

C: The interval scale uses numerical data on the dependent variable where the differences between numbers are assumed to be equal. The scale itself is arbitrary but differences can be meaningfully compared. Example: rating a restaurant on food quality from 0-10. While the numbers themselves are arbitrary, the difference between a 4 and 5 rating is the same as an 8 to 9 rating. No true zero and “twice as much” does not exist. With an interval scale, you can change the range from let’s say 0-5 to 0-10.

A

C: The interval scale uses numerical data on the dependent variable where the differences between numbers are assumed to be equal. The scale itself is arbitrary but differences can be meaningfully compared. Example: rating a restaurant on food quality from 0-10. While the numbers themselves are arbitrary, the difference between a 4 and 5 rating is the same as an 8 to 9 rating. No true zero and “twice as much” does not exist. With an interval scale, you can change the range from let’s say 0-5 to 0-10.

25
Q

Describe the ratio scale of measurement:

A: The ratio scale of measurement involves arbitrary numerical values, similar to the interval scale, without indicating true absence or allowing meaningful multiples.

B: The ratio scale provides the most information, where the numerical measurement is not arbitrary (counting something concrete) and zero indicates a true absence. On a ratio scale, numbers can meaningfully indicate multiples - for example, if one child hits others twice as many times as another child. Ratio scales allow comparisons like “twice as much” that are not possible on other scales. In the ratio scale, you cannot change ranges.

C: On the ratio scale, zero indicates a true absence, but the numbers themselves are arbitrary and do not allow for meaningful comparisons like “twice as much” or multiples.

A

B: The ratio scale provides the most information, where the numerical measurement is not arbitrary (counting something concrete) and zero indicates a true absence. On a ratio scale, numbers can meaningfully indicate multiples - for example, if one child hits others twice as many times as another child. Ratio scales allow comparisons like “twice as much” that are not possible on other scales. In the ratio scale, you cannot change ranges.

26
Q

What are two things we should be concerned with in our true experiment?

A: Reliability and validity

B: Randomness and generalizability

C: Participant expectations and experimenter bias

A

A: Reliability and validity

27
Q

Describe reliability:

A:

B: Reliability is consistency - specifically, the consistency of a measure or study. Reliability does not necessarily mean accuracy or validity, just that the results are predictably the same. Examples: a friend who is always one hour late, we may not like it but it’s consistent and reliable.

C:

A

B: Reliability is consistency - specifically, the consistency of a measure or study. Reliability does not necessarily mean accuracy or validity, just that the results are predictably the same. Examples: a friend who is always one hour late, we may not like it but it’s consistent and reliable.

28
Q

What are the three different types of reliability?

A: Test-retest Reliability, Interrater Reliability, and Replicability.

B:

C:

A

A: Test-retest Reliability, Interrater Reliability, and Replicability.

29
Q

Describe test-retest reliability:

A:

B:

C: The test-retest reliability examines consistency in a person’s scores when they are retested (dependent variable). For example, giving someone a drug to lower blood pressure and expecting the same change in blood pressure each time they take it. Or taking a personality test today and again in a few days and getting similar results. The key is consistency for the same subject over multiple test administrations.

A

C: The test-retest reliability examines consistency in a person’s scores when they are retested (dependent variable). For example, giving someone a drug to lower blood pressure and expecting the same change in blood pressure each time they take it. Or taking a personality test today and again in a few days and getting similar results. The key is consistency for the same subject over multiple test administrations.

30
Q

Describe interrater reliability:

A: Interrater reliability examines consistency between ratings given by multiple judges (let’s say researchers) observing the same behaviors or performances. For example, if judges in the Olympics consistently gave similar scores to figure skating routines this would be an example of high interrater reliability which indicates the ratings are not dependent on the particular judge, but reflect an objective assessment of what was observed. However, if two judges gave the skater a 10 and one gave the skater a 5 this would be an example of low interrater reliability which indicates the ratings are dependent on a particular judge.

B:

C:

A

A: Interrater reliability examines consistency between ratings given by multiple judges (let’s say researchers) observing the same behaviors or performances. For example, if judges in the Olympics consistently gave similar scores to figure skating routines this would be an example of high interrater reliability which indicates the ratings are not dependent on the particular judge, but reflect an objective assessment of what was observed. However, if two judges gave the skater a 10 and one gave the skater a 5 this would be an example of low interrater reliability which indicates the ratings are dependent on a particular judge.

31
Q

Describe replicability reliability:

A:

B: Replicability reliability is the consistency of study results when other researchers conduct additional studies testing the same hypothesis. If multiple studies cannot replicate the original findings, the original conclusions may be discounted. Replicability ensures the effects are not specific to just the original research conditions and can be observed independently by other investigators.

C:

A

B: Replicability reliability is the consistency of study results when other researchers conduct additional studies testing the same hypothesis. If multiple studies cannot replicate the original findings, the original conclusions may be discounted. Replicability ensures the effects are not specific to just the original research conditions and can be observed independently by other investigators.

32
Q

Describe validity:

A: Validity examines whether a measure or study is accurately assessing what it is intended to measure. Specifically, the professor contrasted validity with reliability by noting reliability is about consistency but does not necessarily mean something is accurate or valid. Validity looks at whether a measure or study design is truly capturing the constructs they aim to assess. Example: suppose you have a watch that stops working at 10 minutes past 1:00, this is a reliable watch because it is consistent and always shows that it’s 10 minutes past 1:00 but it is not accurate because it’s not always 10 minutes past 1:00.

B:

C:

A

A: Validity examines whether a measure or study is accurately assessing what it is intended to measure. Specifically, the professor contrasted validity with reliability by noting reliability is about consistency but does not necessarily mean something is accurate or valid. Validity looks at whether a measure or study design is truly capturing the constructs they aim to assess. Example: suppose you have a watch that stops working at 10 minutes past 1:00, this is a reliable watch because it is consistent and always shows that it’s 10 minutes past 1:00 but it is not accurate because it’s not always 10 minutes past 1:00.

33
Q

What are the six different types of validity?

A:

B:

C: Construct (which includes criterion, convergent, and discriminant), internal, and external

A

C: Construct (which includes criterion, convergent, and discriminant), internal, and external

34
Q

Describe construct validity:

A: Construct validity examines whether an operational definition accurately reflects and captures the intended construct it aims to measure. It asks, “is your operational definition appropriate?” It has three components: criterion validity, convergent validity, and discriminant validity. High construct validity means the operational definition accurately measures the construct and not other unintended things. Example: number of yawns is not a good measure of intelligence, (low construct validity), but the results of a test might be a better measure (high construct validity).

B:

C:

A

A: Construct validity examines whether an operational definition accurately reflects and captures the intended construct it aims to measure. It asks, “is your operational definition appropriate?” It has three components: criterion validity, convergent validity, and discriminant validity. High construct validity means the operational definition accurately measures the construct and not other unintended things. Example: number of yawns is not a good measure of intelligence, (low construct validity), but the results of a test might be a better measure (high construct validity).

35
Q

Describe criterion validity:

A:

B: Criterion validity (a component of construct validity) examines whether an operational definition correlates with a concrete behavior it is supposed to represent. For example, whether SAT scores accurately predict and correlate with students’ actual first-year college performance, as the SAT claims to measure readiness for college. The answer is no, it doesn’t, which means that SAT scores are an example of low construct validity for first-year college performance. Criterion validity assesses if a measure relates to real-world outcomes it aims to forecast.

C:

A

B: Criterion validity (a component of construct validity) examines whether an operational definition correlates with a concrete behavior it is supposed to represent. For example, whether SAT scores accurately predict and correlate with students’ actual first-year college performance, as the SAT claims to measure readiness for college. The answer is no, it doesn’t, which means that SAT scores are an example of low construct validity for first-year college performance. Criterion validity assesses if a measure relates to real-world outcomes it aims to forecast.

36
Q

Describe convergent validity:

A:

B:

C: Convergent validity examines whether a measure correlates with other measures that are shown to assess the same construct. For example, if a new extraversion test is supposed to measure extraversion, does it correlate with other validated tests of extraversion? High convergent validity provides evidence an operational definition is accurately capturing the intended construct.

A

C: Convergent validity examines whether a measure correlates with other measures that are shown to assess the same construct. For example, if a new extraversion test is supposed to measure extraversion, does it correlate with other validated tests of extraversion? High convergent validity provides evidence an operational definition is accurately capturing the intended construct.

37
Q

Describe discriminant validity:

A:

B:

C: Discriminant validity examines whether a measure correlates more strongly with what it aims to assess versus other unrelated constructs. It helps ensure a measure is specifically honing in on the intended construct and not also reflecting or being influenced by unrelated factors. For example, an extraversion test should measure extraversion and not self-esteem.

A

C: Discriminant validity examines whether a measure correlates more strongly with what it aims to assess versus other unrelated constructs. It helps ensure a measure is specifically honing in on the intended construct and not also reflecting or being influenced by unrelated factors. For example, an extraversion test should measure extraversion and not self-esteem.

38
Q

Describe internal validity:

A: Internal validity checking the integrity of the relationship between the independent and dependent variables in an experiment. It assesses whether changes in the dependent variable can truly be attributed to the manipulation of the independent variable, rather than being caused by other factors. High internal validity means the independent variable is likely the cause of changes in the dependent variable. She gave the example of yoga (IV) influencing happiness (DV). You would want to make sure that it was actually the yoga that was influencing happiness and not an extraneous variable like soothing music or socializing for example. Establish interval validity in a controlled setting.

B:

C:

A

A: Internal validity checking the integrity of the relationship between the independent and dependent variables in an experiment. It assesses whether changes in the dependent variable can truly be attributed to the manipulation of the independent variable, rather than being caused by other factors. High internal validity means the independent variable is likely the cause of changes in the dependent variable. She gave the example of yoga (IV) influencing happiness (DV). You would want to make sure that it was actually the yoga that was influencing happiness and not an extraneous variable like soothing music or socializing for example. Establish interval validity in a controlled setting.

39
Q

Describe external validity:

A:

B: External validity examines whether the results of a study can be generalized more broadly - specifically, whether the findings generalize to other participants, materials, settings, or times beyond just those used in the study. Higher external validity means effects are more likely to extend beyond the specific sample, procedures, and conditions of the original research. For example, if we said that yoga causes happiness can we generalize this to all exercise? However, we cannot generalize too broadly beyond what we tested. The Mozart study is a good example of generalizing too broadly. External validity will be higher if you include a more diverse population. Establish external validity with additional experiments.

C:

A

B: External validity examines whether the results of a study can be generalized more broadly - specifically, whether the findings generalize to other participants, materials, settings, or times beyond just those used in the study. Higher external validity means effects are more likely to extend beyond the specific sample, procedures, and conditions of the original research. For example, if we said that yoga causes happiness can we generalize this to all exercise? However, we cannot generalize too broadly beyond what we tested. The Mozart study is a good example of generalizing too broadly. External validity will be higher if you include a more diverse population. Establish external validity with additional experiments.