Week 1 - Descriptive Statistics Part One Flashcards
Learn content week one
What is variability?
Variability represents the difference between individuals in some characteristics eg blood pressure, height, reaction to exposure to infections.
What are the two main reasons for collecting data?
First reason = for descriptive purposes and second for comparative purposes.
Define a descriptive study and give an example.
Descriptive statistics include vital statistics eg. the statistics of a population size, births and deaths. Descriptive studies find the answer to these statistics. An example of this would be a study which determines the length of survival of patients diagnosed with colorectal cancer in a certain hospital. Descriptive studies don’t try to understand why.
Define comparative studies and give an example.
Comparative studies are carried out to answer questions. For example is 6 months chemotherapy a satisfactory treatment for pulmonary tuberculosis compared with and established 9 month course of treatment?
Define a Quantitative variable.
A variable that can be counted or measured and involves numbers
What are the different kinds of quantitative variables?
Discrete variable and continuous variable.
A discrete variable is…
A quantitative variable whose possible values are integers (whole numbers) eq. number of children, number of visits to GP. In other words you can not have 1 and a half children.
A continuous variable is….
a quantitative variable that has an uninterrupted range of values eg blood pressure, weight etc. in other words the number can have decimal points.
What is a Qualitative Variable?
A variable that is not able to be counted or characterized by a numerical quantity. Eg sex, occupation, housing, diet etc.
What are the different kinds of qualitative variables?
Ordinal variables, nominal variables, dichotomous variables.
An ordinal variable is…..
A qualitative variable with several characteristics which can be ordered eg type of diet may be rated as poor, good, excellent
A nominal variable is….
A Qualitative variable with categories which have no natural order and can not be placed in an order. Eg disease category may include heart disease, cancer, respiratory disease etc – these are all diseases and have no natural rank or order.
A Dichotomous variable is….
A qualitative variable with just two classes eg yes/no, male/ female, dead/ alive, symptoms absent/ present. This is essentially a special version of a nominal variable.
Define frequency and the different ways frequency is presented tables/ diagrams.
Frequency refers to the number of individuals with a certain characteristic or within a certain range of a variable. Frequency can be presented as the actual number (frequency) or as relative frequency, cumulative frequency and cumulative relative frequency.
calculate the relative frequency, cumulative frequency and cumulative relative frequency in the below example.
calculate the relative frequency in the below example.
answer below
calculate the relative frequency and the cumulative relative frequency for the below example
answer attached
fill out the table below to represent the distribution of parity for each age group. What is important to note about two way tables?
it is important to note when looking at two way tables whether you are calculative column percentages or row percentages. Do not present both in the same table. the below is an example of calculating only the row percentages
what is the name of the below chart and what variables can be plotted using these charts?
Bar Chart. Bar Charts are used to plot quantitative discrete or qualitative variables.
what is the name of the attached chart and how does it differ to a bar chart?
Histogram.
Historgrams are used to display frequency distributions of continuous variables. the values of the variable are given on the horizontal scale and the frequency (or relative frequency) on the vertical scale. the area of the rectangle above each group is proportional to the frequency.
what are other forms of bar charts that can be used to present data and what are the positives and negatives of these kinds of bar charts. what kind of bar chart is attached.
multiple bar charts (bars are side by side) or composite bar charts (bars stacked on each other)
Multiple bar chart presents two different sets of data side by side that make comparisons of data simple.
composite bar charts are difficult to interpret due their set up and are therefore less helpful unless you are interested in the total of all categories.
The bar chart attached is a multiple bar chart.
What is the below graph called and what is it useful for?
cumulative relative frequency plot. shows cumulative (relative) frequency plotted against the variable. in the prev example it is helpful for estimating the median age by drawing a horizonal line from the 50% mark until it hits the distribution curve then dropping it to the horizonal axis
when reporting percentages, how should your numbers be rounded up?
When reporting percentages for a sample size of less than 100 you should not use decimal places as they imply unwarranted precision. In larger sample sizes one decimal place can be used but it is often not needed. See attached for further rounding guidelines.
What is crucial to follow when presenting results in a table or graph?
It is important that tables and figures are prepared according to guidelines (see attached)
