Measurement Scales
What measurement scale is eye color?
Nominal
Can a measure be reliable but not valid, why or why not?
Reliability assesses consistency and validity assesses the extent to which the measure relates to the construct
A research finds that peoples' scores on a measure of extroversion stay stable over 2 months? What type of reliability or validity have they established?
Test-retest reliability
Which sampling techniques create the least biased samples?
Simple random sampling
Systematic random sampling
What type of claim:
Eating kale increases one's weight loss.
Causal
What measurement scale is a 10-point well-being scale?
Interval
What is similar about face and content validity?
They are both subjective ratings
What type of reliability would need to be assessed? A researcher places unobtrusive video recording devices in the hallway of a local high school. Later, coders viewed videotapes and coded how many students were using cell phones in the 4-minute periods between classes.
Interrater
A political polling company wants to get an estimate of how many voters in Wisconsin will vote for each candidate. Which sampling method will more likely result in a representative sample of voters?
A) The company obtains a list of all phone numbers of Wisconsin residents and calls every 5th number on the list?
B) The political polling company picks 5 districts that lean towards one party and 5 districts that lean towards the other party. The company goes to public places in those districts and asks people who are willing to talk to them which candidate the person is willing to voting for?
Option A because it uses random sampling
Sleep is associated with a reduction in coffee intake. What criterion for causation does this claim appear to support?
Covariance
A meteorologist measures the weather on the top and the bottom of a mountain in degrees Celsius. What measurement scale is the researcher using?
Interval
A new self-report measure of extroversion correlates with an observation of the number of conversations each person has in a day. What validity does this self-report measure demonstrated?
Criterion
A researcher samples 50 people and gives each of them a five item scale that to assesses extroversion. Two of the items do not correlate with the other items. What has the researcher not established and why might the items not correlate with each other?
Internal reliability
The items might be assessing a different construct
What is a key difference between stratified random sampling and quota sampling?
One is a random sampling measure and the other non-random sampling
How can a researcher establish temporal precedence with a correlation study?
Longitudinal study
A researcher has 1/2 the participants watch a violent video game and the other 1/2 watch a neutral video game. The participants are then given a punching bag and the researcher measures how many times the participants in each group punch the bag. What measurement scale?
Ratio
What is the difference between internal reliability and internal validity?
Interval reliability- consistency between items
Internal validity- no third variable
A shorter version of a traditional IQ test was developed. Experts in the field said that the new IQ test had all the elements of an IQ test. However, participants who took the new IQ test two times performed very differently each time. What aspects of reliability and/or validity did the researchers establish and what aspects did they not establish?
Content validity-established
Test-retest reliability- not established
What are some problems in Psychology research in terms of external validity?
Mostly test WEIRD sample
Not use random sampling techniques
Two different studies tried to replicate the same effect. Both studies had an r value around .35. However, one study was significant and one study was not significant. What might be different about the two studies?
Precision of the estimate: variability and/or sample size
A researcher asks all participants what is their highest level of education. Then the researcher attempts to run statistics on this data. Which analyses can the researcher not conduct from the education level data: mean, median, range, mode, and confidence interval. Explain your rationale.
Mean and confidence interval because the data is ordinal and there is not equal intervals.
A professor gives a class of 40 people his five-item measure of conscientiousness (e.g., “I get chores done right away,” “I follow a schedule,” “I do not make a mess of things”). Average scores are more highly correlated with a self-report measure of tidiness (r = .50) than with a measure of general knowledge (r = .09). Which two validities has the professor established?
Convergent and discriminate
A group of researchers developed a new method for diagnosing children with autism. It is a survey that asks parents to rate a variety of different behaviors that their child engages in. What are three ways in which the researcher should proceed in terms of validating this measure?
Criterion validity- does it correlate with behavioral diagnosis of autism
Convergent validity- does it correlate with other autism scales
Divergent validity- does it not correlate with ADHD
How could a larger sample be less externally valid than a smaller sample?
If the larger sample did not use random sampling techniques.
A researcher found a correlation of r = .75 with only 10 participants. What questions might you have with this data?
Is there an outlier that is driving the results?