Measurements and Their Analysis
Population
- Consists of all possible observations that can be made on a particular item.
- Thus can not be practically observed.
- Often a population has an infinite number of observations
- The mean and variance of a population are its true values
Sample
- Subset of data collected from a population
- Can be collected in an economical fashion
- May or may not be representative of a population
- Size of same set influences its ability to predict true values of a population.
- Larger samples generally have a higher probability of predicting true values.
True Value
- A quantities theoretically correct or exact value.
- True values are never known
- Often we observe a calibration value with a more precise instrument using very precise techniques. These values are the used as true values with standard equipment.
Most Probable Value
- The most likely value for the true value as computed from a sample data set.
- For a simple set of repeated observations, this is the mean (average value).
Redundancies
- Also called degrees of freedom
- The number of observations in excess of the number necessary to solve for the unknowns.
How many redundancies are their if a distance is observed three times?
Methods of Analyzing Data
- Numerical methods
- Range: The difference between the largest and smallest values in a set of data
- That is, the largest discrepancy in a set of data
- Also called dispersion.
- Discrepancy: The difference between two elements in a set of data.
- Measures of central tendency
- Mean, median, mode
- Measures of data variation
- Variance, standard deviation
- Graphical methods
- Frequency distribution: A bar graph based on a selected class width.
- Also called a histogram
- Class: A sub-region of data
- Class width: range of a single class
- Median: The physical midpoint of a data set when the data is arranged in numerical order
- If the data set has an odd number of elements, then it is the physical midpoint
- If the data set has an even number of elements, then it is the average of the two points nearest to the physical midpoint.
Creating a Frequency Distribution
1. Select the number of classes
- Typically between 5 and 10
2. Determine the class width
- Class width = Range/(number of classes)
3. Create a class frequency table
- Determine the number of elements in each class
4. Plot frequency of each class; i.e., create histogram.
Things To Look For
- Notice the following when looking at a histogram
- Symmetry about central values
- Range of data set
- Frequency of occurrence of measured values
- Steepness of histogram
- Measure of precision
- The steeper the slopes of the sides are the more precise the data.
- How to Lie with a bar graph
- Note lowest value in histogram
- When comparing two data sets, they must use the same class width