Measurement Error Confounds Flashcards
Define Random error
A chance fluctuations in our measurement/ error by chance
Define Constant or systematic error
A bias that is present and influences our
measurement in a systematic way
Random errors (bias/obscure) the results
Constant errors (bias/obscure) the results
Random errors (obscure) the results
Constant errors (bias) the results
What variables are undesirable variables that add error to our experiments?
Extraneous variables
What are Extraneous variables?
Undesirable variables that add error
to our experiments
What is the aim of research when it comes to Extraneous variables?
Aim = To eliminate or at least control the
influence of extraneous variables
Methods = Random allocation/counterbalancing which spreads the influence of extraneous variables
Define internal validity
The degree of confidence that the causal relationship being tested is trustworthy and not influenced by other factors or variables
Simply = Is it valid in the lab/as a lab experiment?
Define external validity
The extent to which you can generalise the findings of a study to other situations, people, settings, and measures
Simply = Can the results be applied/related to real-world settings?
Define Confounding Variables
A 3rd variable in a study examining a potential cause and effect
It may mask or distort the effects of another variable
What do Confounding variables measure?
1) An effect of the IV on the DV when it is not present
2) An effect of the IV on the DV when it is present
What are the 4 main sources of confounding variables/threat to internal validity?
1) Selection
2) History
3) Maturation
4) Instrumentation
How does “selection” cause a threat to internal validity?
Selection = Bias resulting from the selection or assignment of participants to different levels of the IV
e.g. Putting Psych students in 1 group and students from a different course with no prior knowledge of Psych in another group
Consequences = When you allocate Ps into groups unfairly but controlled selection and not random allocation, it could affect the measurement of the DV
Why does selection error happen often in quasi-experiments?
Because quasi-experiments are unable to randomly allocate participants
(low internal validity but high external validity)
How does “history” cause a threat to internal validity?
History = Uncontrolled events that take place between testing occasions
e.g. Testing a group of people with alcohol in the morning and another group with alcohol in the afternoon
There could be a change that occurs during the day that makes the testing situation different (maybe more traffic noise in the afternoon, more distracting)
How does “maturation” cause a threat to internal validity?
Maturation = Intrinsic changes in the characteristics of participants between
different test occasions
e.g. In within-subjected design, participants get older and more mature which may affect their performance
Why does maturation error happen often in within-subjects experiments?
Because WS is a longitudinal study (study participants over time) and Ps may have matured a lot since the last testing session
How does “instrumentation” cause a threat to internal validity?
Instrumentation = Changes in the sensitivity or reliability of measurement instruments during the course of the study
e.g. Some measuring devices may become less sensitive when they deteriorate and may affect measurements of DV
What is Reactivity in Validity?
When Ps or experimenters are aware that they are being observed and they may alter their behaviour
What are the 2 types of reactivity?
1) Subject-related reactivity
2) Experimenter related
What is Subject-related reactivity?
When participants behave how they think the researchers want them to behave (they may be aware of the aims of the experiment and which experimental condition they are under)
What is Experimenter-related reactivity?
When experimenters give inadvertent verbal/non-verbal cues to the participants that could bias the results (they may be aware of which experimental condition the Ps are under)
How do you overcome reactivity?
Blind procedures
What are blind procedures?
Information which may influence the participants and/or experimenters of the experiment is withheld until after the experiment is complete
What is a single and what is a double blind procedure?
Single = When only participants are unaware of the IVs of the study in which they are operating in
Double = When both the participants and experimenters are unaware of the IVs of the study in which the participants are operating in
You are measuring the height of women and make small errors reading the tape measure. Are these errors…
a. Random b. Constant
A
You forget to ask the women to take their shoes off before you measure them. Are these measures…
a. Random b. Constant
B
You weigh participants on a set of scales that are not set to zero. Are these errors…
a. Random b. Constant
B
Which of the following is not a common threat to internal validity?
a. Maturation b. Iteration c. Selection d. History
B
Which of the following statements is correct?
a. Random errors obscure the results, Constant errors bias the results b. Constant errors obscure the results, Random errors bias the results c. Constant errors tend to push the measurements up and down around the exact value d. Random errors tend to push the measurements in the same direction
A
Intrinsic changes in the characteristics of participants between different test occasions is a threat to internal validity as a result of…
a. Selection b. History c. Maturation d. Instrumentation
C
You set out to determine the effect of mobile phone use on hazard awareness while driving. On the first occasion, you ask participants to identify potential hazards during a driving simulation with no distracter. On the second occasion, you ask the same participants to repeat the task while talking on a mobile phone.
a. Solution to confound
b. Subjects design
c. Confound
d. Dependent variable
e. Hypothesis
f. Independent variable
- Talking on a mobile will reduce drivers’ ability to identify hazards
- Mobile use- with versus without
- Number of distractors identified
- Within-subjects/ Repeated measures
- Maturation: order effects (practice/fatigue)
- Counterbalance conditions across participants
- E
- F
- D
- B
- C
- A
You want to determine whether the introduction of social media tools has made children feel more or less isolated. You compare children’s ratings of loneliness in 2011 with children’s ratings of loneliness in 1991.
a. Confound
b. Subjects design
c. Independent variable
d. Hypothesis
e. Dependent variable
- Children will obtain higher/lower scores on loneliness measures following the introduction of social media
- Pre/post social media 2011/1991
- Score on loneliness measure
- Between/independent
- History: other changes that could impact the DV (e.g. rise in mobile phone use/ email communication).
- D
- C
- E
- B
- A
