Scientific Inquiry, Data Analysis, and Laboratory Skills
Scientific Inquiry, Data Analysis, and Laboratory Skills tests experimental design, measurement quality, graphing, basic statistics, and common bench techniques. Questions often present a scenario and ask what conclusion is supported, what control is missing, or how to calculate a concentration or colony count.
Core rules
- Design: identify the independent variable, dependent variable, and controlled variables. Include a negative control, a positive control when relevant, randomization, and replication. Blind or double-blind procedures reduce bias.
- Accuracy and precision: accuracy is closeness to the true value; precision is closeness among repeated measurements. Random error affects precision; systematic error affects accuracy.
- Statistics: standard deviation describes spread of data; standard error of the mean (SEM = SD/√n) describes uncertainty in a mean. Error bars must be identified before interpretation. A t-test compares two means; chi-square compares observed and expected frequencies.
- Graphs: use bar graphs for categorical comparisons and scatter or line graphs for continuous relationships. Label axes with units and show variation.
- Bench skills: spectrophotometry requires a blank; centrifugation separates by density; gel electrophoresis separates DNA by size; serial dilutions multiply dilution factors; sterile technique prevents contamination.
Worked example
A 0.1 mL sample from a 10⁻⁴ dilution is plated, and 150 colonies grow. Colony-forming units per mL = 150/(0.1 × 10⁻⁴) = 150/10⁻⁵ = 1.5 × 10⁷ CFU mL⁻¹. The reasoning: dividing by the plated volume converts colonies to CFU per mL of the diluted sample, and dividing by the dilution factor converts back to the original concentration. If the plate had 30 colonies or 300, the count would be less reliable because the acceptable range is usually about 30–300 colonies.
Traps
- Confusing technical replicates, repeated measurements of one sample, with biological replicates, independent samples.
- Reporting SEM as if it were SD, which understates variability when n is large.
- Forgetting to subtract the blank in spectrophotometry or to convert units before calculating concentration.
- Treating correlation as causation, especially in observational field data.
- Miscounting serial dilutions by adding rather than multiplying dilution factors.