Email: support@essaywriterpros.com
Call Us: US - +1 845 478 5244 | UK - +44 20 7193 7850 | AUS - +61 2 8005 4826

# the Organizational Development website

Fundamentals of Test Evaluation

Lesson 19 – Analysis and Evaluation RESOURCES I PRIMT I HELP

Basic Analytic Techniques

A variety of analy tic techniques are used during the statistical analysis of test data. Using these techniques effectively for test design and later for data analysis requires significant levels of mathematical training as well as practical experience. Some of the more common techniques used for statistical inference are summarized in simplified form below:

Hypot hesis Test ing: Setting up and testing a hypotheses is an essential part of statistical inference. In order to formulate such a test, usually some theory has been put forward, either because it is believed to be true or because it is to be used as a basis for argument, but has not been proved. For example, claiming that the SRAW warhead arming distance is 16 meters.

As part of Hypothesis Testing, two hypotheses are set up. One is typically referred to as the “null” hypothesis (e.g., SRAW will arm in less than 16 meters); the other is the “alternative” hypotheses (e.g., SRAW will not arm is less than 16 meters) . Sophisticated mathematical techniques are applied to determine degree and risks of rejection or acceptance of these hypotheses.

T Test : the T-test assesses whether the means of two groups are statistically different from each other and is used in hypothesis testing under certain circumstances. This test is appropriate whenever you want to analy tically compare the means of two groups. For example, if the difference between the mean values of measured noise levels from testing on two different production lots of the SRAW are statistically significant. The T test is used most often on small sample sizes .

….rl I Page8of 24 ~ Back Next

TST102 Fundamentals of Test Evaluation

Lesson 19 – Analysis and Evaluation RESOURCES I PRIMT I HELP

Basic Analytic Techniques, Cont.

ANOVA: Analysis of variance is a collection of statistical models, and their associated procedures, in which the observed variance is partitioned into components due to different explanatory variables. ANOVA is used in place of the T test when an experiment or test involves more than two samples.

Factorial Analysis: Is a technique to determine effects of many test factors, such as operator, production lot, weather, etc. simultaneously during a test event in the most optimum way . This technique is one that is used in test design to get the maximum amount of valid data from the minimum amount of tests . Done properly, a single test can produce data that can be used to assess multiple criteria.

Regression Analysis: Involves plotting data in a manner so that mathematical analysis can be performed on it to determine “best fit” curves that account for the relationships between variables. For instance, performing a regression analysis of data related to SRAW accuracy at 800 meters as a function of gunner engagement angles. A graphic relationship between the two, based on actual test data that depicts all possible engagement angles would be developed. Accuracy for SRAW engagement angles not explicitly tested can then be estimated by interpolation or extrapolation from the graphic .

Interpolation is a moderately risky method of estimating untested values within the data set. Ex trapolation is estimating values by extending the regression line beyond the data set and is high risk because grossly inaccurate predictions may result.