Limits...
Common Statistical Pitfalls in Basic Science Research

View Article: PubMed Central - PubMed

AUTOMATICALLY GENERATED EXCERPT
Please rate it.

The analysis of clinical samples, population samples, and controlled trials is typically subjected to rigorous statistical review... Although determining an appropriate sample size for basic science research might be more challenging than for clinical research, it is still important for planning, analysis, and ethical considerations... In basic science research, there is often no prior study, or great uncertainty exists regarding the expected variability of the outcome measure, making sample size calculations a challenge... A significant statistical finding (eg, P<0.05 when the significance criterion is set at 5%) is due to a true effect or a difference or to a type I error... A type I error is also known as a false‐positive result and occurs when the hypothesis is rejected, leading the investigator to conclude that there is an effect when there is actually none... Conversely, a comparison that fails to reach statistical significance is caused by either no true effect or a type II error... A type II error is described as a false‐negative result and occurs when the test fails to detect an effect that actually exists... Minimizing type II error and increasing statistical power are generally achieved with appropriately large sample sizes (calculated based on expected variability)... A common pitfall in basic science studies is a sample size that is too small to robustly detect or exclude meaningful effects, thereby compromising study conclusions... In designing even basic science experiments, investigators must pay careful attention to control groups (conditions), randomization, blinding, and replication... The goal is to ensure that bias (systematic errors introduced in the conduct, analysis, or interpretation of study results) and confounding (distortions of effect caused by other factors) are minimized to produce valid estimates of effect... It is common to find basic science studies that neglect this distinction, often to the detriment of the investigation because a repeated‐measures design is a very good way to account for innate biological variability between experimental units and often is more likely to detect treatment differences than analysis of independent events... Investigators often design careful studies with repeated measurements over time, only to ignore the repeated nature of the data with analyses performed at each time point... Such an approach not only fails to examine longitudinal effects contained in the data but also results in decreased statistical power compared with a repeated‐measures analysis... Survival analyses can be particularly challenging for investigators in basic science research because small samples may not result in sufficient numbers of events (eg, deaths) to perform meaningful analysis.

No MeSH data available.


Related in: MedlinePlus

Blood flow over time by strain. *P<0.05.
© Copyright Policy - creativeCommonsBy-nc
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5121512&req=5

jah31794-fig-0005: Blood flow over time by strain. *P<0.05.

Mentions: We wish to compare organ blood flow recovery over time after arterial occlusion in 2 different strains of mice. The outcome of interest is again normalized blood flow (a continuous outcome), and the comparison of interest is the trajectory (pattern over time) of mean normalized blood flow between strains. The unit of analysis is the mouse, and we have repeated measurements of blood flow (before occlusion, at the time of occlusion [time 0], and then at 1, 3, 7, 14, 21, and 28 days). Data can be summarized as shown in Figure 5, in which means and standard error bars are shown for each time point and compared statistically using repeated‐measures ANOVA (again, assuming that normalized blood flow is approximately normally distributed). Note that analyses at each time point would not have addressed the main study question and would have resulted in a loss of statistical power.


Common Statistical Pitfalls in Basic Science Research
Blood flow over time by strain. *P<0.05.
© Copyright Policy - creativeCommonsBy-nc
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5121512&req=5

jah31794-fig-0005: Blood flow over time by strain. *P<0.05.
Mentions: We wish to compare organ blood flow recovery over time after arterial occlusion in 2 different strains of mice. The outcome of interest is again normalized blood flow (a continuous outcome), and the comparison of interest is the trajectory (pattern over time) of mean normalized blood flow between strains. The unit of analysis is the mouse, and we have repeated measurements of blood flow (before occlusion, at the time of occlusion [time 0], and then at 1, 3, 7, 14, 21, and 28 days). Data can be summarized as shown in Figure 5, in which means and standard error bars are shown for each time point and compared statistically using repeated‐measures ANOVA (again, assuming that normalized blood flow is approximately normally distributed). Note that analyses at each time point would not have addressed the main study question and would have resulted in a loss of statistical power.

View Article: PubMed Central - PubMed

AUTOMATICALLY GENERATED EXCERPT
Please rate it.

The analysis of clinical samples, population samples, and controlled trials is typically subjected to rigorous statistical review... Although determining an appropriate sample size for basic science research might be more challenging than for clinical research, it is still important for planning, analysis, and ethical considerations... In basic science research, there is often no prior study, or great uncertainty exists regarding the expected variability of the outcome measure, making sample size calculations a challenge... A significant statistical finding (eg, P<0.05 when the significance criterion is set at 5%) is due to a true effect or a difference or to a type I error... A type I error is also known as a false‐positive result and occurs when the hypothesis is rejected, leading the investigator to conclude that there is an effect when there is actually none... Conversely, a comparison that fails to reach statistical significance is caused by either no true effect or a type II error... A type II error is described as a false‐negative result and occurs when the test fails to detect an effect that actually exists... Minimizing type II error and increasing statistical power are generally achieved with appropriately large sample sizes (calculated based on expected variability)... A common pitfall in basic science studies is a sample size that is too small to robustly detect or exclude meaningful effects, thereby compromising study conclusions... In designing even basic science experiments, investigators must pay careful attention to control groups (conditions), randomization, blinding, and replication... The goal is to ensure that bias (systematic errors introduced in the conduct, analysis, or interpretation of study results) and confounding (distortions of effect caused by other factors) are minimized to produce valid estimates of effect... It is common to find basic science studies that neglect this distinction, often to the detriment of the investigation because a repeated‐measures design is a very good way to account for innate biological variability between experimental units and often is more likely to detect treatment differences than analysis of independent events... Investigators often design careful studies with repeated measurements over time, only to ignore the repeated nature of the data with analyses performed at each time point... Such an approach not only fails to examine longitudinal effects contained in the data but also results in decreased statistical power compared with a repeated‐measures analysis... Survival analyses can be particularly challenging for investigators in basic science research because small samples may not result in sufficient numbers of events (eg, deaths) to perform meaningful analysis.

No MeSH data available.


Related in: MedlinePlus