November 6, 2018

# Understanding Research For Clinical Simulation, Part 3: Introduction to Statistics

Simulation staff need to understand research theory either to fully participate in their own department’s research or to review the current literature with a view to improving their own simulation program. Today in Part 3 of HealthySim’s Series on “Understanding Research”, we introduce some of the key basics of statistics. Linked below are all the parts of the exclusive healthcare simulation series!

An Intro to Statistics

Definition:  the practice or science of collecting and analyzing numerical data in large quantities, especially for the purpose of inferring proportions in a whole from those in a representative sample (Oxford Dictionary). Stats are used to collect, organize, present, analyze, interpret numerical data for the purpose of making more effective decisions.

• Terminology:
• Descriptive statistics – summative data that has been collected – often made into charts.
• Inferential statistics – to generalize from a sample population to population as a whole.
• Testing hypotheses
• Population – collection of all possible individuals, objects or measurements of interest.
• A sample –  a part of or portion of the population of interest.
• Levels/Scales of Measurement:
• Nominal Level  – naming items such as ethnicity, social security number, religion.
• May be classified  into categories.
• Categories are mutually exclusive.
• Exhaustive – each individual, object or item appears in at least one category.
• Cannot be averaged.
• E.g. Pt dietary preference ( vegan, vegetarian etc), patient medical record number, gender, hair color, blood group
• Ordinal Level – items are ranked.  One category is higher than another.
• Each category has higher order e.g. ranks in the army or levels of disease category.
• E.g. Sim experience rating :  Very good, good, satisfactory, unsatisfactory, very unsatisfactory.
• Interval Level –  the distance between levels e.g. time, age (not categories).
• All the characteristics of nominal and ordinal scale of measurement are met.
• The zero point is arbitrary e.g. temperature starting at 93.0’F
• E.g. Interval levels on a thermometer – Fahrenheit temperature or Centigrade temperature.
• Ratio Level – highest level of measurement
• All requirements of the interval scale are met except the zero on the scale does means “does not exist”.
• Zero point is meaningful.
• Ratio of numbers is meaningful
• E.g. \$0 in your pocket versus -\$20 which has no meaning.
• E.g. Distance travelled is 20 or 10 miles.  Twenty miles is twice as far.
• Temperature is not a ratio scale because zero temperature exists

How Percentages Work in Stats

• Percent means per hundred.
• For example 75% is 75 out of 100
• Alternatively, 75% may be considered as ¾
• To find 75% of a number multiply by 75 and divide by 100 – this is the same as multiplying by 0.75
• 75% of 364 = 365 x 75, then divide by 100 = 273 or
• 364 x 0.75 = 273
• FYI IV fluids are often written as a percentage.
• E.g. D5% 0.9%NaCl
• Where the percent stands for number of grams of a solute in 100 mL of solution
• D5 = Dextrose 5% or 5 grams of dextrose in 100 mL solvent (water).
• 0.9% NaCl = 0.9 grams salt (sodium chloride) in 100 mL or 9 grams in 1000 mL or one liter.
• To calculate the percentage divide the number of items or patients or students by the total population or group.
• E.g.  60 out of 80 students thought simulation was beneficial
• 60/80 x 100 = 75%.  The total in the class was 80.
• Percentages are a favorite way for people to misrepresent information. Having the actual data provides more information.
• E.g.  3 out of 4 students (75%) thought simulation beneficial but this is a very low sample number (n=4).   Clearly, 60 out of 80 students provides more reliable information since the sample size is so much bigger.

Measure of Central Tendency – Mean, Median and Mode

• Mean (aka average)
• Add up all the values in a set of data and then divide by the number of values added.
• E.g.  Average age in a class of 11 students:
• Ages of 11 students: 23, 23, 23, 24, 25, 26, 27,29, 29, 32, 40
• Total = 269/11 = Average age of class 24.5 years.
• Median (Mdn)
• The score point at or below which 50 percent of the values fall.
• In the case above, the value of 26 has 5 students below the age of 26 and 5 students with an age greater than 26.
• The advantage of using median if the data has one or two very large or very small numbers.
• E.g. Age of group – 20, 21, 21, 22, 40
• Median 21
• Average = 124/5 = 24.8
• Mode (Mo)
• In ungrouped data, the mode is the score that appears most frequently.
• In the class above, 23 appears 3 times and is the most frequent.
• There may be more than one mode in a set of data.
• Might be useful when ordering equipment size.
 Normal Ordinal Interval Ratio Mean X X Yes Yes Median X Yes Yes Yes Mode Yes Yes Yes Yes

The above chart beaks down the key differences between Mean, Median and Mode.

Understanding Research for Clinical Simulation Series: