Week Two Flashcards
Define inferences
Draw conclusions about parameters based on statistics
What is sampling error
The difference between the population parameter and the sample statistics. Need to keep sampling error as low as possible
How do you reduce sampling error
Need some random selection of participants into the sample
Randomness is crucial to probability sampling techniques
Equality of the draw; muse be equally likely to be selected
Independence of the draw: selecting one element won’t affect the sampling of any other elements
What are the three measures of central tendency
Mean: arthimetic average (total of all observed values divided by the number of values)
Median: mid point value which has 50% of ranked or ordered data fall above or below
Mode: most common or frequently observed value
What level of measurements are associated with the measures of central tendency?
Mean: interval or ratio
Median: ordinal
Mode: nominal or categories data
When should you use each measure of central tendency
Mean: if variable is numeric and doesn’t have extreme scores
Median: if variable is numeric and has extreme scores must be at least ordinal
Mode: if variable is categorical and modal category is observed
What is the normal curve
Shows the distributions of our data
The normal curve is symmetrical, no skewnesss, can be Kurtose, unimodal so only one peak
Mean, mode and medium all the same under normal curve
Bimodal: two peaks
What is skewness
Concerned w the symmetry of the distribution
Skewed distribution have one side of the distribution that is different to the other
Positive skew: the tail of the graph points towards the positive end of the scale
Negative skew: the tail of the graph points toward the negative end
What is the difference between a positive and negative skew?
Negative: elongated tail at the left.
Positive: enlongated tail at the right
When is symmetry considered to be obtained with skewness
When the skewnesss value is -1 to 1 - aiming for zero
What is kurtosis
The extent to which cases are piled up around the measure of central tendency
What are the three difference types of kurtosis peak ness
Leptokurtic: values are all close to the measure of central tendency and variance and std dev smaller (lept - tall)
Platykurtic: values are spread across the distribution and variance and std dev largest (platypus - flat)
Mesokurtic: values close to normal distribution (middle- normal)