曝光臺 注意防騙 網曝天貓店富美金盛家居專營店坑蒙拐騙欺詐消費者

dependent variables.
􀂃 Need to select appropriate experimental design
and experimental environment.
􀂃 Examples:
– Compare the usability of the Macintosh OS and Win 98 for
novice and expert users.
– Evaluate the control room layout of a nuclear power plant.
– Evaluate the ATC control tower layout.
– Evaluate the control (helm) performance of different
cockpit.
A project supported by AIRBUS and the CAAC
A project supported by AIRBUS and the CAAC
June 2007
Procedures for doing research
1. Determine what project you are going to do.
2. Collect and review related literature: Go to
library, use on-line search to locate related
journal articles, then read the articles and
summarize them.
3. From summarized information from literature
review and what you've learned from class, build
a model to represent your project.
4. Discuss and modify your model.
5. Choose an experimental environment. (In lab or
in the field)
A project supported by AIRBUS and the CAAC
June 2007
Procedures for doing research
6. Determine the independent and dependent
variables for your model (or criterion variables
and stratification variables if it's a descriptive
study).
7. Determine appropriate experimental design for
validating your model.
8. Choose subjects.
9. Collect data.
A project supported by AIRBUS and the CAAC
A project supported by AIRBUS and the CAAC
June 2007
Procedures for doing research
10 Analyze data with adequate statistics.
􀂃 Choose appropriate statistical model to analyze
data:
􀂾 Simple regression model
􀂾 Multiple regression model
􀂾 Analysis of variance model (ANOVA)
􀂾 Multivariate analysis of variance model (MANOVA)
􀂾 Nonparametric procedures
􀂃 Choose samples
􀂾 Representative sample
11 Report your results.
A project supported by AIRBUS and the CAAC
June 2007
About "Model"?
􀀴 A model is something more abstract than real
system to represent the system under study
with appropriate details to describe parts of
interest whereas simplify the parts not in the
scope of interest.
􀀴 Example: Human-machine system.
􀀴 SHEL model
􀀴 Reason model
􀀴 HFACS
A project supported by AIRBUS and the CAAC
A project supported by AIRBUS and the CAAC
June 2007
Research Criterion
􀀴 Criterion Measures in Research
􀀴 Practical requirements
􀀴 Reliability
􀀴 Validity
􀀴 Sensitivity
A project supported by AIRBUS and the CAAC
June 2007
Research Criterion
Criterion Measures in Research
􀂃 Characteristics and behaviors measured in
descriptive studies
􀂃 Dependent variables in the experimental studies
􀂃 Basis for judging the goodness of a design in
evaluation research.
A project supported by AIRBUS and the CAAC
A project supported by AIRBUS and the CAAC
June 2007
Research Criterion
􀀴 Practical Requirements
􀂃 Objectivity
􀂃 Quantitative
􀂃 Unobtrusive
􀂃 Easy to collect
􀂃 Require no special data collection techniques or
instrumentation
􀂃 Cost as little as possible in terms of:
􀂾 time
􀂾 experimenter energy, and/or
􀂾 money
A project supported by AIRBUS and the CAAC
June 2007
Research Criterion
􀀴 Reliability
􀂃 Consistency and stability of the measures of the
variable(s) over timer or across representative
samples.
A project supported by AIRBUS and the CAAC
A project supported by AIRBUS and the CAAC
June 2007
Research Criterion
􀀴 Validity
􀂃 Face Validity
􀂃 Content Validity
􀂃 Construct Validity
A project supported by AIRBUS and the CAAC
June 2007
Research Criterion
􀀴 Validity
􀂃 Face Validity
􀂾 The measure looks like valid.
􀂾 Why is this important?
– If it doesn’t look as though it is valid – the subjects of the
research may not give their full cooperation.
– If the subjects don’t cooperate fully, the research will fail.
– Therefore, in this case, appearance is vital.
􀂃 Content Validity
􀂃 Construct Validity
A project supported by AIRBUS and the CAAC
A project supported by AIRBUS and the CAAC
　
中國航空網 www.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：AIRWORTHINESS AND FLIGHT STANDARDS(30)