2.4 Basic Statistics and Analysis Flashcards
What is the difference between descriptive and inferential statistics?
A: Descriptive statistics provide data about the past (mean, standard deviation, range, number of readings) and inferential Statistics is used to help explain the future (regression analysis, hypothesis testing, confidence intervals, and time series analysis).
What is included in a Minitab descriptive statistics summary report?
A: Descriptive statistics (normality, central tendency, dispersion, and confidence intervals for key statistics), Histogram and box plot, and comparison of confidence intervals for mean and median.
What are mean, median, and mode (Central Tendency descriptive stats)?
A: Mean is the average, Median is the middle point of observations (resistant to outliers), and mode is the most common observation.
What are the dispersion descriptive statistics?
A: Range, Standard Deviation, Interquartile Range, and Coefficient of Variation.
What is the definition of 1st Quartile and 3rd Quartile?
A: 1st quartile is drawn at 25% of the data and the 3rd quartile is drawn at 75% the data.
What is the IQR and why is it important?
A: The IQR is Q3 – Q1. The IQR * 1.5 above or below Q3 or Q1 would be create the whiskers. Observations beyond the whiskers are considered outliers and marked with a dot or x.
What is a Histogram?
A: A histogram is a specialized bar graph that shows how often certain observations occur. The width of the bar, often called a bin, is determined by the observations it contains. The height of the bin is determined by the frequency of observations in the data that fit in the bin. The size of the bin depends on what you want the data to show.
What is a normal curve?
A: The normal curve is the shape that tends to appear when we observe a measurement is affected by random factors. The Central Limit Theorem tells us that even a non-normal sample is likely to normalize when many samples are taken. Because we know the dimensions of the normal curve, we can start to use it to make predictions and calculate probabilities.
What is skewness?
A: Is a measurement of how symmetrically the data is distributed. Positive skew is >0 skewed right, Mean > Median and Negative skew is < 0, Mean < Median, skewed left. Skew usually has an assignable cause
What is a Boxplot?
A chart that shows the first, second and third quartiles of a data set. It also uses whiskers beyond the quartiles to show outliers. The length of the whiskers is usually (1.5*IQR).
What is the Coefficient of Variation (CV) and what is it used for?
The standard deviation divided by the mean. It is a quick way to see how big the standard deviation is.
Generally, it follows the same rules as MSA (10% good, 30% ok, 30%+ bad).
It is especially useful when comparing the variation of several different datasets.
What is the Standard Error of the Mean and what is it used for?
The standard deviation divided by the square root of the number of data points (𝑠/√𝑛). It is key in calculating the confidence intervals for the population mean. The larger the standard error of the mean is, the less sure we are about the true mean of the population.
