peč H

INTRODUCTION TO BASIC CONCEPTS IN EXPERIMENTAL DESIGN

a ODNO ——

1.3

Sampling Distribution. A theoretical probability distribution that describes the
functional relation between possible values of a statistic based on N cases drawn
at random and the probability associated with each value over all possible samples
of size N.

id

Statistical Model. A mathematical statement concerning the sampling dis-
tribution of a random variable that is used in evaluating the outcome of an experi-
ment or in predicting the outcome of future replications of an experiment.

Test Statistic. A statistic whose purpose is to provide a test of some statistical
hypothesis. Test statistics such as t and F have known sampling distributions that
can be employed in determining the probability of an obtained result under the
null hypothesis. |

Relative Efficiency of a Statistic. Ratio of experimental error of one statistic
to that of another statistic.

Statistical Decision Theory. Branch of mathematics concerned with the
problem of decision making and the choice of decision rules under uncertain
conditions.

FORMULATION OF PLANS FOR
COLLECTION AND ANALYSIS OF DATA

ACCEPTABLE RESEARCH HYPOTHESES

Some guestions cannot currently be subjected to experimental
investigation. For example, the guestions "Can three or more angels sit
on the head of a pin?" and "Does life exist in more than one galaxy in the
universe? cannot be answered because no procedures presently exist for
observing either angels or other galaxies. Scientists confine their research
hypotheses to guestions for which procedures can be devised that offer
the possibility of arrivingat an answer. This does not mean that the guestion
concerning the existence of life on other galaxies can never be investigated.
Indeed, with continuing advances in space science it is probable that this
guestion will eventually be answered.

Ouestions that provide the impetus for experimental research should
be statable in the logical form of the general implication. That is, a guestion
should be reducible to the form, if A, then B. For example, if albino rats
are subjected to microwave radiation, then their food consumption will
decrease. This research hypothesis can be investigated because procedures
both for manipulating radiation level and for measuring food consumption
of rats are available.

DISTINCTION BETWEEN DEPENDENT AND
INDEPENDENT VARIABLES

In the example just cited, the presence or absence of radiation is
designated as the independent variable —the variable that is under the