Part 1. Chapter 3, page 57-85 Three Claims, Four Validities
Variables
Variables = something that varies, so it must have at least 2 levels or values.
Constant = something that could potentially vary but that has only one level in the study in
question.
à Example: ‘’smoking’’ would be a variable, and its levels would be smoker and nonsmoker
Measured variable = one whose levels are simply observed and recorded. Measuring a
variable is a matter of recording an observation, a statement, or a value as it occurs
naturally.
à IQ can be measured, such as age, gender, hair color.
Manipulated variable = variable a researcher controls, usually by assigning study
participants to the different levels of that variable.
à a researcher might give some participants 10mg of medication, others 20mg and still
others 30mg. Or a researcher might assign some people to take a test in a room with many
other people and assign others to take the test alone.
Conceptual variables = are abstract concepts, such as ‘’spending time socializing’’ and
‘’school achievement’’. Sometimes called a construct.
- Must be carefully defined at the theoretical level à these definitions called
conceptual definitions.
Operational definitions (operational variables) = to test the hypothesis the researchers
have to do something specific in order to gather data. They do so by testing their hypothesis
with empirical data.
A claim = an argument someone is trying to make.
Three kinds of claims:
1. Frequency claims = describe a particular rate or degree of a single variable (two out
of five)
- They focus on only one variable, such as frequency of worrying, rate of smiling, or
amount of texting.
- Frequency variables are always measured not manipulated!
2. Association claims = argues that one level of a variable is likely to be associated with
a particular level of another variable.
- They sometimes said to correlate = that when one variable changes, the other
tends to change too. à correlation study
Positive association = association in which high goes with high and low goes with low.
Negative association = high goes with low and low goes with high
Zero association = there is no association between variables
Making predictions based on Associations
Some association claims are useful because the help us make predictions. Which couples are
going to stay together the longest? If we know the level of one variable, we can more
accurately guess, or predict, the level of the other variable.
1
, • The stronger the relationship between 2 variables, the more accurate our prediction
will be; the weaker the relationship between 2 variables, the less accurate our
prediction will be.
• But if 2 variables are even somewhat correlated, it helps us make better predictions
that if we didn’t know about this association
à both positive and negative associations can help us predict. Zero associations cannot.
3. Causal claims = arguing that one of the variables I responsible for changing the other.
- Example: those who take music lessons have higher IQ
- Occasionally you might also see causal claim based on a zero association à a lack of
cause.
- Example: vaccines do not cause autism.
à words like cause, enhance, affect and change suggest that there is (no) causal
claim.
- Causal claims are a step above
association claims. Because
they make a stronger
statement, we hold them to
higher standards.
To move from a simple language of
association to the language of causality
a study has to satisfy 3 criteria:
1. The causal variable and outcome
variable are correlated (relationship
cannot be zero)
2. Causal variable came first and
outcome variable came later
3. No other explanations exist for the
relationship.
Interrogating the three claims using
the four big validities
1. Construct validity
2. External validity
3. Statistical Validity
4. Internal validity
à Validity refers to the appropriateness of a conclusion or decision, and in general, a valid
claim is reasonable, accurate and justifiable.
Interrogating Frequency Claims
To evaluate how well a study supports a frequency claim, you will focus on 2 of the big
validities: construct validity and external validity.
Construct validity = how well a conceptual variable is operationalized. When evaluating the
construct validity of a frequency claim, the question is how well the researchers measured
2
, their variables. Example: claim is: 4 in 10 teens admit to texting while driving. There are
several ways to measure this variable.
à asking teenagers to tell you on an online survey
à using phone records
à in other words: there are several ways to operationalize such a variable
When you ask how well a study measured or manipulated a variable, you are interrogating
the construct validity: how accurately a researcher has operationalized each variable.
à using the accurate, most fitting measurement (meetinstrument)
To ensure construct validity, researchers must establish that each variable has been
measured reliably à meaning the measure yields similar scores on repeated testing) and
that different levels of variable accurately correspond to true differences in, say, depression
or happiness.
External validity of Frequency Claims
The next questions to ask about frequency claims concern generalizability: How did the
researchers choose the study’s participants, and how well do those participants represent
the intended population? This is about external validity
External validity = how well the results of a study generalize to, or represent, people or
contexts besides those in the original study.
Statistical Validity of Frequency Claims
Statistical validity (statistical conclusion validity) = the extent to which a study’s statistical
conclusions are accurate and reasonable. How well do the numbers support the claim?
The percentage reported in a frequency claim is usually accompanied by a margin of error of
the estimate = statistical figure based on simple size for the study, that attempts to include
the true value in the population.
à in the report about how many teenagers text while driving, the Centers for Disease
Control’s 41% value was accompanied by this note. ‘’The margin of error is + or – 2.6%
percentage points. à the margin of error helps us describe how well our sample estimates
the true percentage. Specifically the range, 38,4%-43,6%, is high likely to contain the true
percentage of teens who text while driving.
Interrogating Association Claims
Studies that are able to support association claims are called correlational studies: they
measure 2 variables instead of just one.
Construct validity of association claims
To support an association claim, a researcher measures 2 variables so you have to assess the
construct validity of each variable.
Example: people who multitask the most are the worst at it.
1. You should ask how well the researchers measured the frequency of multitasking and how
well they measured the ability to multitask.
3
