UCSB Comm 88 Final Exam with
Complete Solutions
Multiple time series design Correct Ans-two time series designs, one with treatment, one
without. tests whether treatment caused change
Within subjects design Correct Ans-every subject is in every condition. compare same group
of subjects instead of different conditions. individual responses per condition. subjects are their
own control group. show how subjects react to different conditions
Carry over/Order effects Correct Ans-the order in which the manipulation was done is what
was producing the results. EX) practice effects, fatigue effects, first treatment/contrast effects
Practice effects Correct Ans-done something once, affects it later. prepared
Fatigue effects Correct Ans-tired of experiment, done too many responses
First treatment/contrast effect Correct Ans-whatever you do first contrasts the next
Solutions to carry over/order effects Correct Ans-counterbalance orders - randomly assign
order of treatments to each subject
, Strengths of Laboratory experiments Correct Ans-highly controlled setting = high internal
validity and ability to make sure everything is the same across all conditions
Weaknesses of Laboratory experiments Correct Ans-artificial setting = low external validity,
must watch more carefully for experiment effects and reactivity effects
Strengths of Field experiments Correct Ans-natural setting/behavior = high external validity,
less reactivity - subjects don't know they're being watched
Weakness of Field experiments Correct Ans-harder to maintain experimental control
Content analysis (quantitative) Correct Ans-systematically, quantitatively examining the
content of communication
Uses of content analysis Correct Ans-describe how much/what kind of messages there are,
asses "image" of groups in media, compare media to real world, examine message changes over
time, provide background on media effects, code/analyze open ended data in
surveys/experiments
Content analysis sampling Correct Ans-define population of interest, identify "unit of
analysis" for coding, select representative sample (ideally)
Complete Solutions
Multiple time series design Correct Ans-two time series designs, one with treatment, one
without. tests whether treatment caused change
Within subjects design Correct Ans-every subject is in every condition. compare same group
of subjects instead of different conditions. individual responses per condition. subjects are their
own control group. show how subjects react to different conditions
Carry over/Order effects Correct Ans-the order in which the manipulation was done is what
was producing the results. EX) practice effects, fatigue effects, first treatment/contrast effects
Practice effects Correct Ans-done something once, affects it later. prepared
Fatigue effects Correct Ans-tired of experiment, done too many responses
First treatment/contrast effect Correct Ans-whatever you do first contrasts the next
Solutions to carry over/order effects Correct Ans-counterbalance orders - randomly assign
order of treatments to each subject
, Strengths of Laboratory experiments Correct Ans-highly controlled setting = high internal
validity and ability to make sure everything is the same across all conditions
Weaknesses of Laboratory experiments Correct Ans-artificial setting = low external validity,
must watch more carefully for experiment effects and reactivity effects
Strengths of Field experiments Correct Ans-natural setting/behavior = high external validity,
less reactivity - subjects don't know they're being watched
Weakness of Field experiments Correct Ans-harder to maintain experimental control
Content analysis (quantitative) Correct Ans-systematically, quantitatively examining the
content of communication
Uses of content analysis Correct Ans-describe how much/what kind of messages there are,
asses "image" of groups in media, compare media to real world, examine message changes over
time, provide background on media effects, code/analyze open ended data in
surveys/experiments
Content analysis sampling Correct Ans-define population of interest, identify "unit of
analysis" for coding, select representative sample (ideally)
