Deck 23: Data Science and Analytics in Healthcare

Question 1

Which measure is most important for evaluating the performance of classifiers, models that predict class membership?

A) Root mean squared error (RMSE)
B) Area under the receiver operating characteristic (ROC) curve
C) Percentage of correctly classified cases
D) Hosmer-Lemeshow statistic

Answer 1

Area under the receiver operating characteristic (ROC) curve

Question 2

Which specialized method is used in knowledge discovery and data mining?

A) Stratified sampling
B) Conceptual analytic methods
C) Selection of the optimal subset of variables/features for use in modeling
D) Machine learning

Answer 2

Selection of the optimal subset of variables/features for use in modeling

Question 3

The process of knowledge discovery and data mining is best characterized as:

A) machine learning, use of statistical methods, and large amounts of data.
B) a process wherein patients are selected for clinical trials.
C) "fishing" for information in data.
D) cleaning data for further analysis using other methods.

Answer 3

machine learning, use of statistical methods, and large amounts of data.

Question 4

As organizations mature over time in their use of data science, they progress from what to what?

A) Retrospective analysis; exploratory data analyses
B) Real-time analytics; retrospective analytics
C) Exploratory data analyses; prescriptive analytics
D) Data mining; knowledge discovery

Question 5

A health analytics professional must be skilled in all of the following except:

A) computer hacking.
B) healthcare operations.
C) management and economics.
D) statistical understanding.

Question 6

Assigning individual records or rows in a dataset for specific purposes (eg, model development, model validation) is a process known as:

A) partitioning.
B) sampling.
C) data mining.
D) Bayesian networking.

Question 7

Which methods are examples of data mining? (Select all that apply.)

A) Decision trees
B) Stratified random sampling
C) Bayesian networks
D) Artificial neural networks
E) Factor analysis

Question 8

Which of the following are examples of analytics tools? (Select all that apply.)

A) Graphics programs like Photoshop and 3DS Max
B) Spreadsheets like MS Excel
C) Database management systems like SQL Server
D) Statistical analysis programs like Rapidminer
E) Programming languages like R or Python

Question 9

In a set of consistently coded clinical data, the data should be analyzed using descriptive statistics and visualization with respect to the following: (Select all that apply.)

A) Distribution
B) Arcing
C) Sparsity
D) Identifiers
E) Outliers

Question 10

Which of the following data characteristics are important in understanding big data in healthcare? (Select all that apply.)

A) Variety
B) Vagrancy
C) Venality
D) Velocity
E) Volume