Scientific Inquiry & Reasoning Skills - Skill 4: Data-based Statistical Reasoning

Like questions about Skill 3, questions that test Skill 4 will ask you to show you can “do” science, this time by demonstrating your data-based and statistical reasoning skills. Questions that test this skill will ask you to reason with data. They will ask you to read and interpret results using tables, graphs, and charts. These questions will ask you to demonstrate you can identify patterns in data and draw conclusions from evidence.

Questions that test this skill may ask you to demonstrate your knowledge of the ways natural, behavioral, and social scientists use measures of central tendency and dispersion to describe their data. These questions may ask you to demonstrate your understanding of the ways scientists think about random and systematic errors in their experiments and datasets. They may also ask you to demonstrate your understanding of how scientists think about uncertainty and the implications of uncertainty for statistical testing and the inferences they can draw from their data. These questions may ask you to show how scientists use data to make comparisons between variables or explain relationships between them or make predictions. They may ask you to use data to answer research questions or draw conclusions.

These questions may ask you to demonstrate your knowledge of the ways scientists draw inferences from their results about associations between variables or causal relationships between them. Questions that test this skill may ask you to examine evidence from a scientific study and point out statements that go beyond the evidence. Or they may ask you to suggest alternative explanations for the same data.

For example, questions that test this skill will ask you to use your knowledge of data-based and statistical reasoning by:

  • Using, analyzing, and interpreting data in figures, graphs, and tables to draw a conclusion about expected results if the experiment was to be completed again.
  • Evaluating whether representations are an appropriate or reliable fit for particular scientific observations and data.
  • Using measures of central tendency (mean, median, and mode) and measures of dispersion (range, inter-quartile range, and standard deviation) to describe data.
  • Using reasoning about random and systematic error.
  • Using reasoning about statistical significance and uncertainty (e.g., interpreting statistical significance levels, interpreting a confidence interval) and relating this information to conclusions that can or cannot be made about the study.
  • Using data to explain relationships between variables.
  • Using data to answer research questions and draw conclusions.
  • Identifying conclusions supported by research results.
  • Determining the implications of results for real-world situations.
  • Using structural comparisons to make predictions about chemical properties in an unfamiliar scenario.

For example, questions from the Psychological, Social, and Biological Foundations of Behavior section may ask you to demonstrate your use of data-based and statistical reasoning by:

  • Identifying the correlation between a demographic variable, such as race/ethnicity, gender, or age, and life expectancy or another health outcome.
  • Identifying the relationship between demographic variables and health variables reported in a table or figure.
  • Explaining why income data are usually reported using the median rather than the mean.
  • Using reasoning to identify or evaluate what inference is supported by a table of correlations between different socioeconomic variables and level of participation in different physical activities.
  • Using reasoning about the type of comparisons made in an experimental study of cognitive dissonance and evaluating what the findings imply for attitude and behavior change.
  • Drawing conclusions about the type of memory affected by an experimental manipulation when you are shown a graph of findings from a memory experiment.
  • Distinguishing the kinds of claims that can be made when using longitudinal data, cross-sectional data, or experimental data in studies of social interaction.
  • Identifying which conclusion about mathematical understanding in young children is supported by time data reported in a developmental study.
  • Evaluating data collected from different types of research studies, such as comparing results from a qualitative study of mechanisms for coping with stress with results from a quantitative study of social support networks.
  • Using data, such as interviews with cancer patients or a national survey of health behaviors, to determine a practical application based on a study’s results.

The three questions that follow illustrate Skill 4 questions from, respectively, the Psychological, Social, and Biological Foundations of Behavior section; the Biological and Biochemical Foundations of Living Systems section; and the Chemical and Physical Foundations of Biological Systems section of the MCAT exam.

Skill 4 Example From the Psychological, Social, and Biological Foundations of Behavior Section

Which correlation supports the bystander effect?

  1. The number of bystanders is positively correlated with the time it takes for someone to offer help in the case of an emergency.
  2. The number of bystanders is negatively correlated with the time it takes for someone to offer help in the case of an emergency.
  3. The number of bystanders is positively correlated with whether people judge a situation to be an emergency.
  4. The number of bystanders is negatively correlated with whether people judge a situation to be an emergency.

The correct answer is A. This Skill 4 question assesses knowledge of Content Category 7B, Social processes that influence human behavior. It is a Skill 4 question because it requires you to engage in statistical reasoning. This question requires you to understand the distinction between negative and positive correlations and make a prediction about data based on your knowledge of the bystander effect.

Skill 4 Example From the Biological and Biochemical Foundations of Living Systems Section

In the figure, the three curves represent hemoglobin oxygen binding at three different pH values, pH 7.2, pH 7.4, and pH 7.6.

 

unnamed file

What conclusion can be drawn from these data about the oxygen binding of hemoglobin at different pH values?

  1. Low pH favors the high-affinity oxygen-binding state.
  2. Low pH favors the low-affinity oxygen-binding state.
  3. Oxygen affinity is independent of pH.
  4. Oxygen binding is noncooperative at low pH.

The correct answer is B. This Skill 4 question draws on knowledge from Content Category 1A, Structure and function of proteins and their constituent amino acids. This is a Skill 4 question because it asks you to use data to explain a property of hemoglobin. You must evaluate the hemoglobin oxygen-binding data for each pH value and compare them to determine the relationship between pH and hemoglobin oxygen affinity in order to conclude that low pH favors the low-affinity oxygen-binding state.

Skill 4 Example From the Chemical and Physical Foundations of Biological Systems Section

Four different solutions of a single amino acid were titrated, and the pK values of the solute were determined.

 

Solution

pK1

pK2

pK3

1

2.10

3.86

9.82

2

2.10

4.07

9.47

3

2.32

9.76

Not Applicable

4

2.18

9.04

12.48

Which solution contains an amino acid that would be most likely to stabilize an anionic substrate in an enzyme pocket at physiological pH?

A. Solution 1

B. Solution 2

C. Solution 3

D. Solution 4

The correct answer is D. This Skill 4 question includes a table and assesses knowledge of Content Category 5D, Structure, function, and reactivity of biologically relevant molecules. Here you see that four different solutions of a single amino acid were titrated, and the pK values were determined. These values are found in the table. This is a Skill 4 question because you must recognize a data pattern in the table, make comparisons, and use those comparisons to make a prediction. Using knowledge of amino acids and peptide bonds and the patterns you see in the data, you can determine that the N- and C-terminus pK values, roughly 2 and 9 for all solutions, can be ignored since these groups will be involved in peptide bond formation. With further analyses, you can determine that only Solution 4 will be cationic at physiological pH.