Statistical Methods Flashcards
• data
• statistical methods
• art
• science
• can neither prove nor disprove anything
• it is just a tool
• discipline concerned with the treatment of numerical data derived from group of individuals
Statistics
• branch of statistics
• applied to biological or medical science
Biostatistics
Biostatistics various steps
generation of hypothesis
collection of data
application of statistical analysis
• know the data
• it’s distribution
• it’s analysis
Biostatistics
In order to draw a valid conclusion
Biostatistics
2 branches of Statistical Method
Descriptive Statistics
Inferential Statistics
Distribution of population measurements by providing types of data, estimates of central tendency, measures of variability
Descriptive Statistics
3 Types Central Tendency
mean, mode, median
Two Types of Variability
standard deviation, correlation coefficient
Used to express the level of certainly about estimates includes hypothesis testing, standard error of mean, confidence interval
Inferential statistics
Three Types of Data
Nominal
Ordinal
Interval
Observation recorded during research constitute
Data
Simply assigned “names” or categories
Nominal Data
Based on presence or absence of certain attributes/characteristics
Nominal Data
Without any ranking between the categories
Nominal Data
It also includes binomial data, which refers to two possible outcomes
Nominal Data
Expressed as scores or ranks. Also called as ordered, categorical or graded data
Ordinal Data
Characterized by equal and definite interval between 2 measurements
Interval Data
Types of Interval Data
Continuous
Discrete
Can take any value within a given range
Continuous Data
Usually assigned integer values i.e does not have fractional values
Discrete Data
Average, Common measure of central tendency, Most widely used in calculations of averages, Least affected by sampling fluctuations
Mean
Middle value, Used for scores and ranks, Best indicator of Central Value
Median
Most frequent value, the point of maximum concentration, Used when values are widely varying, Rarely used in medical studies
Mode
Tells you on average how much any given observation is different from the mean
Standard Deviation
It tells you on average how far each score lies from the mean
Standard Deviation
In normal distributions, a standard deviation means that values are generally far from the mean
High Standard Deviation
In normal distributions, a standard deviation indicates that value are clustered close to the mean
Low Standard w
Describes the Variability of observation about the mean
Describes the _____ of the population
Measures the ______ of a data distribution
Measures the _______ between each data point and the mean
Standard Deviation
Scatter
Spread
Typical Distance
To calculate SD, we need it square called
Variance
The average square deviation around the mean
Variance
We divide by the number of data points
N
We divide by one fewer than the number of data points
N-1
Measure the degree of linear relationship between two continuous variables
Correlation Coefficient
To understand whether one data (such as a person’s weight) can help predict another data variable (such as the same person’s blood pressure) via a linear relationship
Correlation Coefficient
To assess the strength of associations between data variables
Correlation Coefficient
A correlation coefficient is a number between
-1 and 1
Tells you the strength and direction of a relationship between variables. It reflects how similar the measurements of two or more variables are across a dataset
Correlation Coefficient
Can provide insights into complex real-world relationship. Helping researchers develop theories and make predictions
Correlation research
The maximum value of 1 is obtained if there is a straight line is scatter plot and considered as perfect
Positive correlation
The association is _____ if the the values of x-axis and y-axis tend to be high or low together
Positive
The association is ______ ie. -1 if the high y-axis values tends to go with low values of x-axis
Negative
Every positive increase in one variable, there is a proportional positive increase in the other variable. For instance, belt size increase almost perfectly in correlation with waist size
1 = Strong Positive Linear Relationship
Every positive increase in one variable, there is a proportional negative decrease in other variable
-1 = Strong Negative Linear Relationship
No linear relationship between the variables
0
The absolute value of the correlation coefficient represents the _______ of the relationship
Strength
A larger absolute value indicates a stronger relationship, stronger is the _______
association
|-.75| =.75, has a stronger relationship than
.65
A _______ correlation depends on the discipline
meaning
Physics correlation coefficient should be between
-0.95 and 0.95
Social sciences a correlation coefficient should be results between _______ are meaningful
-0.6 and 0.6
Correlation between the two variable does ____ necessarily suggest the _____ and _____ relationship
Not
Cause
Effect
While performing (Correlation) these tests, it requires _____ to be normally distributed. It is generally used to form hypothesis and to suggest areas of future research
x and y variables
There a several types of correlation coefficient, but most popular is
Pearson’s
A correlation coefficient commonly used in linear regression
Pearson’s R
