Study Guides (248,289)
CSB345H1 (9)

Session 1 - Descriptive Statistics.docx

4 Pages
189 Views

Department
Cell and Systems Biology
Course
CSB345H1
Professor
William Navarre
Semester
Fall

Description
HMB325H © Lisa | Page 13 L E C T U R E 1 : H O W T O S U M M A R I Z E D ATA ( C H A P T E R 2 ) LEARNING OBJECTIVES 1. list, recognize, describe the characteristics of the 5 main levels/types of data measurement used in biomedical studies 2. list and describe the descriptive statistics (ex. measures of central tendency, variability) 3. describe the situations & circumstances in which each of these descriptive statistics should & should not be used 4. create & critique graphical displays of data distributions 5. describe the characteristics of normal & skewed distributions 6. describe what the standard of error of the mean measures & the 2 factors that affect its size 7. calculate & interpret applicable confidence intervals THE 3 MAIN TASKS OF STATISTICS 1. to describe/summarize a set of data – data reduction 2. to estimate how close a result might be to the ‘true’ (unobserved) value 3. to estimate the probability that random variation (‘chance’) explains the observed results – inferential statistics 1. hypothetical example on a community health survey: During the past 24 hours, how many cigarettes have you smoked? 1. significance of this question (why?): smoking has understood adverse health effects 2. who should be studied: hope that sample/subgroup is representative of the whole pop (randomly selected) 1. statistics tells us that on avg, the random sample will be representative 3. what is measured/counted?: a number LEVELS OF DATA MEASUREMENT 2. what ‘level/type of data measurement’ is ‘number of cigarettes smoked’? 1. continuous – no breaks bw two numbers (decimals & fractions are infinite bw them) 2. discrete – whole numbers, ex. # of pregnancies, teeth, stairs climbed 1. mostly discrete – depends on how variables are collected from indiv respondents 3. what other ‘levels/types of data measurement’ are there? 3. categorical 1) nominal – categories w names, no natural order, ex. eye colour, blood type, birth country 2) dichotomous (binary) – only 2 choices, ex. have/don’t have the disease (special kind of nominal variable) 3) ordinal – categories are ordered, ex. a 5-point scale DESCRIPTIVE STATISTICS 4. descriptive statistics is used to turn a set of data (ex. of random numbers) into something more meaningful 5. types of descriptive statistics 1. graphical displays – “a picture’s worth a thousand words”, ex. line graph, stem & leaf plot, number line 2. quantitative measures (descriptive statistics) 1) measures of central tendency (mean, median, mode) – ‘representative’ (central) value 1. mean: adding up the values & dividing the sample by the number (count) of values in the sample 2. median: the middle value of a sequential set of data (the 50 percentile – 50% of the values are less than it & 50% are greater) 3. mode: the most frequently occurring value(s) in a set of data HMB325H © Lisa Z| Page 2 2) measures of variability – ‘dispersion/spread’ of the data 4. range: interval/difference from the lowest to the highest value in a set of data (maps out the extremes but not how the data are dispersed bw them) th 5. interquartile range (IQR): the range th the data’s first quartile (25 percentile) & its third quartile (75 percentile) (p/100)(n+1); p = percentile, n = number of data points ex. the 75 percentile of 14 data points would be the (75/100)(14+1) = 11.25 th th th observation; the 75 percentile is the average of the 11 & 12 observation, ex. 34 & 40, (34+40)/2 = 37, IQR = 0 – 37 1. 50 percentile = median = avg bw the 50 data point & the 51 st 6. standard deviation (SD): the ‘average’ distance each data point is from the sample’s mean 6. computing the standard deviation of 5 example data values: 71, 75, 79, 83, 92 1. Compute the mean: 400/5 = 80.0 2. Subtract the mean from each value: -9.0, -5.0, -1.0, 3.0, 12.0 3. Square each of the differences (“squared deviations”): 81.0, 25.0, 1.0, 9.0, 144.0 4. Add the squares & divide by n–1: 260/4 = 65.0 (the “variance”)  divide by n–1 bc any sample will underestimate the true amount of variability for the whole pop (make denominator 1 unit smaller to mak
More Less

Related notes for CSB345H1
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.