Statistics

Five types of sample statistics include sample mean, sample variance, sample standard deviation, sample proportion.

1. Sample mean is the average of all data points in a sample. It is calculated by summing all values in sample and then dividing by number of observations.
2. The sample variance measures the dispersion and spread of data points in sample. It indicates the average of squared differences from sample mean.
3. Sample standard deviation is the square root of the sample variance that provides a measure of dispersion in same units as data.
4. The sample proportion is fraction of the sample that has a certain attribute or characteristics. This is used for estimating population proportion.
5. Sample median is middle value in an ordered list of sample data. When the sample size is even, the median is the average of two middle numbers.

Since a frequency polygon represents the data distribution, by interpreting the areas under curve, it is possible to infer probabilities for some range and interval. Let us take a look at it:

• Understanding the Data: You must be able to comprehend the data represented by frequency polygon. In most cases, x-axis represents data values/intervals and y-axis represents frequency or relative frequency of those values.
• Conversion to relative frequency: If frequency polygon is based on absolute frequencies, first convert it into relative frequency by dividing every frequency by total number of observations.
• Identifying the area of interest: Then you need to determine the range or interval for which you need to find the probabolity. This might be a specific bin or a range of values on x-axis.
• Estimate area under curve: To find the probability of specific range, you need to estimate area under curve for that range. This is possible by approximating area using geometric shapes such as trapezoids and rectangles.

The following are different types of frequency polygons:

• Simple Frequency Polygon: This is a standard form that connects the midpoints of tops of the bars in a histogram with straight lines.
• Relative Frequency Polygon: This type of frequency polygon uses relative frequencies (proportions or percentages) instead of using absolute frequencies. This frequency polygon is used for comparing datasets of different sizes.
• Cumulative Frequency Polygon (Ogives): Ogives are related and they represent cumulative frequencies. These can be used for showing the cumulative distribution of data and are used with frequency polygons.
• Smoothed Frequency Polygon: Rather than connecting points with straight lines, smoothed curves or splines might are used to give a more continuous appearance to data distribution.
• Overlayed Frequency Polygons: Multiple frequency polygons can be overlaid on same graph for comparing different datasets or distributions.

The following points highlight the importance of frequency polygons:

• Frequency polygons are useful for comparing distributions of multiple datasets on the same graph. It becomes easy to visually compare shapes and trends of different datasets by overlaying multiple frequency polygons.
• These use lines to connect points which provide a continous representation of the data. It is easier to see patterns and trends over intervals through frequency polygons.
• They can simplify the visualization of complex data which makes it easy to interpret the overshap and data distribution without distraction of bins or bars.
• Line format of frequency polygon help in identifying trends, patterns and changes in data over time. This is especially useful in time series analysis.
• Frequency polygons visualize measures of central tendency and spread by providing clear view of data distribution.
• These can be used with different types of data and they are not limited by either number of bins of width of intervals. This makes it a flexible tool for data analysis tasks.