Complete the 12 sentences below. Choose the best answer for each one.
1The results are informative but limited ............... they depend on self-reported exposure rather than independently verified behaviour.
Wrong!
"Insofar as" is correct because it introduces the exact respect in which the finding is limited.
2The convenience sample should be treated as ............... because participants were recruited from only two specialist clinics.
Wrong!
"A potential source of bias" is correct because convenience sampling can systematically distort estimates rather than make them more precise.
3Because the design was cross-sectional, the analysis cannot ............... that stress causes absenteeism.
Wrong!
"Establish" is correct because a cross-sectional design can suggest an association but cannot establish a temporal causal effect.
4With only twelve participants in the treatment subgroup, the study remains ............... for detecting modest interaction effects.
Wrong!
"Underpowered" is correct because the sentence refers to a sample size too small to detect modest subgroup effects.
5Given the highly selected eligibility criteria, the findings may not generalise ...............
Wrong!
"To routine clinical practice" is correct because a highly selected trial population limits generalisability beyond the study setting.
6The absence of ............... follow-up prevents any assessment of whether symptoms preceded the reported exposure.
Wrong!
"Longitudinal" is correct because temporality can be examined only when observations track change over time.
7Although age and sex were adjusted for, ............... cannot be ruled out because disease severity was not recorded.
Wrong!
"Residual confounding" is correct because unmeasured severity remains a possible alternative explanation after adjustment for recorded variables.
8The unequal loss to follow-up between groups raises the possibility of ...............
Wrong!
"Attrition bias" is correct because unequal loss to follow-up can distort the comparison between groups.
9Exposure histories were collected after diagnosis, making the estimates vulnerable to errors in ............... recall.
Wrong!
"Retrospective" is correct because asking participants to remember past exposure is vulnerable to recall error.
10As the survey was purely observational, it should not be used to ............... without manipulation or temporal ordering.
Wrong!
"Infer causation" is correct because the sentence explicitly says there was no manipulation or temporal ordering.
11Using an unvalidated translation of the questionnaire threatens ............... rather than merely increasing random error.
Wrong!
"Measurement validity" is correct because an unvalidated translation may fail to measure the intended construct equivalently.
12Because dropout was associated with symptom severity, the analysis is vulnerable to ............... missingness.
Wrong!
"Non-random" is correct because dropout related to symptom severity means missingness depends on the outcome-related variable.
Exercise complete
0 out of 12
correct answers
Answers
- The results are informative but limited insofar as they depend on self-reported exposure rather than independently verified behaviour.
- The convenience sample should be treated as a potential source of bias because participants were recruited from only two specialist clinics.
- Because the design was cross-sectional, the analysis cannot establish that stress causes absenteeism.
- With only twelve participants in the treatment subgroup, the study remains underpowered for detecting modest interaction effects.
- Given the highly selected eligibility criteria, the findings may not generalise to routine clinical practice.
- The absence of longitudinal follow-up prevents any assessment of whether symptoms preceded the reported exposure.
- Although age and sex were adjusted for, residual confounding cannot be ruled out because disease severity was not recorded.
- The unequal loss to follow-up between groups raises the possibility of attrition bias.
- Exposure histories were collected after diagnosis, making the estimates vulnerable to errors in retrospective recall.
- As the survey was purely observational, it should not be used to infer causation without manipulation or temporal ordering.
- Using an unvalidated translation of the questionnaire threatens measurement validity rather than merely increasing random error.
- Because dropout was associated with symptom severity, the analysis is vulnerable to non-random missingness.