Dr. Miroshynk "Exam Outline"

includes her study guide + anything I think is important! Has all 3 of her powerpoints.

What is statistics?

• A science not a branch of mathematics, but uses mathematical models as essential tools

• functions: summarize, organize, present, analyze, interpret

What is biostatistics?

used to interpret the effects of a medical procedure or medicines on LIVING ORGANISMS

• bio= life

What are descriptive statistics?

measurements of central tendency like mean, median, and mode

What is the difference between a variable and data?

variable- a characteristic that is being observed/measured

data- the measured values assigned to the VARIABLE for each individual member of the population

Ex: pain level would be a variable and each patient’s pain level as severe, mild, moderate, etc. would be the data

What are the 4 levels of data measurement?

1. nominal

2. ordinal

3. interval

4. ratio

Which of the 4 levels of data measurement are discrete and which are continuous?

discrete- nominal, ordinal

• discrete ex’s: 5 kids, 96 workers, 3 laptops

  • were counting

continuous- interval, ratio

• continous ex’s: 3.25 kg, 1.32 miles

What is the difference between 4 levels of measurement?

• nominal- data can only be counted/categorized

• ordinal- data can be counted and ranked

• interval- counted, ranked, and added/subtracted

• ratio- data can be counted, ranked, added/subtracted AND multiplied/divided

Rachel’s tips for the 4 levels of measurement:

• Me personally, I HATE how they explain nominal, ordinal, interval, ratio… so I will explain how I was taught this topic back in the day:

• nominal: we can put things into categories…

  • Ex: gender… I can put people into different categories by gender, however I CANNOT rank them in any order, or do any type of adding, multiplying etc… i mean how can you add/ subtract or rank a gender?

• ordinal: we can put think into categories, but now we can RANK them

  • Ex: winners of a race… I can categorize by 1st, 2nd, 3rd, etc. and NOW there’s a ranking system… we know 1st is the winner, etc. etc. HOWEVER, I still cannot add, subtract, multiply or divide… I mean how do you divide winners of a race?

  • Another ex: letter grades… I categorize as well AS RANK. We know A, B, C, D there’s a ranking system as well as I categorized data based on what their grade was.

• interval: we can categorize, rank, and NOW ADD/SUBTRACT

  • this is when the meaning of numbers start getting into play

  • when you think of adding/ subtracting, think of it more as if there is a scale that allows you to assess the difference

    • Ex: Temperature… We know that between 10-20 degrees there is a 10 degree DIFFERENCE and between 50-60 degrees there is a 10 degree difference. By using adding/subtracting we can now further analyze the data. If I were to subtract the 1st and 2nd place winner of a race… what does that mean????

• Ratio- we can categorize, rank, add/subtract and NOW MULTIPLY AND DIVIDE

  • Ex: prices… a $100 dollar book costs twice the amount of a $50 book. That was division.

  • when I think of this one, I also think of these variables having a meaningful zero, before with interval if we had 0 degrees, that just means its cold, but ratio data like height, cannot be zero… you cannot be 0 inches tall.

Examples of each type of measurement level:

• nominal

  • eye color, gender, race

• ordinal

  • pain-scales, level of satisfaction, stage of disease

• interval

  • temperature

• ratio

  • distance, length, weight

How do you calculate mean, median, and mode?

• mean= average

  • add em all up/ divide by the number of data points

• median= middle

  • list the values from lowest to highest and pick the one in the middle

• mode

  • which value occurs most frequently

PRACTICE:

Identify what type of measurement each example is:

1. Heat measured in degrees Celcius

2. Level of satisfaction for a particular breakfast cereal

3. Length of a capsule

4. Favorite color

1. interval

2. ordinal

3. ratio

4. nominal

Calculate the mean, median, and mode for the following data:

2, 4, 5, 6, 4, 7, 5, 4

mean- 4.625

median- 4.5

mode- 4

How do you calculate a range:

max value- min value

How do you calculate IQR:

• to calculate:

  • IQR= Q3- Q1

  • List numbers from lowest—>highest

  • split that down the middle

  • find the median of the first half—> Q1

  • find the median of the second half—>Q3

What are the measures of distribution shape?

• normal distribution

• skewed distribution

What is the difference between normal and skewed distributions?

normal= symmetrical, bell-shaped curve

• mean=median=mode

skewed= asymmetrical, uneven distribution

• mean≠median≠mode

What does incidence measure?

• measures the number of new cases of a disease during a given period

How is incidence proportion (risk) calculated?

What does prevalence measure?

• # of existing cases (new and preexisting) at a particular point in time

Calculate point prevalence:

= # of existing cases(new and preexisting)/ population

Calculate mortality rate:

Sensitivity is also referred to as ________________________.

true positive rate

Specificity is also referred to as ______________________.

true negative rate

___________________ is the test ability to correctly identify individuals WITHOUT disease.

a. sensitivity

b. specificity

b. specificity or true negative rate

• Think: negative= no disease

___________________ is the test ability to correctly identify individuals WITH disease.

a. sensitivity

b. specificity

a. sensitivity or true positive rate

• Think: positive= tested positive for the disease

How do you calculate sensitivity?

How do you calculate specificity?

How do you calculate accuracy?

EXAM TYPE PROBLEM ON SENSITIVITY AND SPECIFICITY!!!

How do you calculate negative and positive predictive values?

What are the common types of visual data representation?

• frequency table

• box and whisker plot

  • used to present ranges of data

• bar chart

  • ordinal data

• histogram

  • continuous data

• pie chart

What are inferential statistics?

use data to make a conclusion

What is the difference between population and samples and statistic and parameter?

population- set of all individuals in the group of interest

• parameter goes with population

sample- a portion of the population selected for study

• statistic goes with sample

What is the main idea of the central limit theorem in statistical analysis?

as sample size increase= distribution will approach NORMAL

What is the difference between null and alternative hypothesis?

null- assumes no difference between the study groups

alternative- assumes there IS a difference between the study groups

parametric vs. nonparametric tests and their selection criterea

parametric- for normal distributed data

• ex: t-tests, ANOVA

nonparametric- for not normal distributed data

• chi-square, etc.

What are type I and type II errors?

Type I

• false positive

• If you reject the null hypothesis and you weren’t supposed to reject it

Type II

• false negative

Alpha vs p-value

• both are probabilities

• p-value < alpha = REJECT NULL

• p-value > alpha = ACCEPT NULL

What is confidence interval and its value?

confidence interval is another measure of statistical significance

CI= 1- alpha

Explain how confidence intervals can be used to test hypotheses:

• Is used to test if our hypothesis is clinically significant so…

  • estimates a range of scores likely to contain the unknown pop parameter

Which one is more important in clinical research—-statistical or clinical significance?

clinical signficance