1 What are some steps that scientists can take in designing

1.) What are some steps that scientists can take in designing an experiment to avoid false negatives?

Solution

In order to perform meaningful statistical analysis, one needs to design the experiment with statistical principles in mind:

Accurately defining the variable and sample space, factors and the level of the factors.

Identifying the type of experiment and using appropriate controls

replications of experiment to create a representative body of data

One must also understand how other scientists have designed experiments in the same field.

Looking into research papers of similar topics and techniques provides a better insight into how to design your experiment.

Few points to look for are:

Variables being investigated

independent and dependent variables

sample size

controls being used

number of replicates of experiments

Taking into account the above points, one gets an idea about the standards in the field of research.

Differentiating between the qualitative and quantitative variables is also important for the experiment design. Quantitative variables are the ones that differ in magnitude (Eg: absorption, median etc) and qualitative variables differ in their types for example gender or marital status.

Use of Controls:

With the help of controls you can evaluate whether the experiment is working on a technical level by giving you a standard to measure against. Controls also guard against factors that might create a bias on the results.

there are four basic types:

Positive control : Known to give a positive result without fail

Negative Control : Known to give a negative result without fail

Control Groups : When the independent variable is set at a preselected level

Controlled variable: Where the one of the variable has a constant value, to check the variation in the independent variable.

Positive controls are the best way to reduce the probability of a false negative result and vice versa.

For example: Checking whether a simple circuit to light an electric bulb works or not. If the circuit itself is damaged, it will always give false negative result whether a perfectly good bulb is used or not. To avoid such false negative results, a positive control must be used, which in this case would be a circuit which works and lights a working electric bulb.