Types of experiments

Overview

Teaching: 0 min
Exercises: 0 min
Questions
  • How are hypotheses generated or tested?

  • What are my experimental units?

Objectives
  • Differentiate between studies conducted to generate and test hypotheses

  • Explain how exploratory and pilot studies can inform experimental research

Before we go any further, we should clarify what we mean by “experiment,” and how we’ll be using the term in the rest of the course. Fry (2014) notes that “a hypothesis is formulated in the basis of observation or theory and tested by experiment” (p. 460). There are two important clauses in this sentence, each referring to a different kind of methodological approach. Let’s unpack this.

A hypothesis is formulated on the basis of observation and theory

Some procedures may be regarded as experimental in the sense that they involve investigating a matter that is to some extent unknown but do not have the purpose of testing a hypothesis. Examples are the recording of growth curves and organ weights in a novel strain as well as many observational or correlation studies. In what might be termed “exploratory experiments,” the primary objective is to generate information on which to build a hypothesis or look for patterns, and some definitions of the scientific method include this as a stage in the process. In “pilot experiments,” the intention is to provide preliminary information that can improve the conduct, quality, and efficiency of subsequent hypothesis-testing experiments, typically using small numbers of experimental units. (Fry, 2014, p. 460, reprinted with permission)

A hypothesis is tested by experiment

Exploratory experiments and pilot studies are intended to study the relationships between variables, or test the feasibility of a research method and/or intervention. These studies do not necessarily employ the strict procedures and multiple study arms seen in a randomized controlled trial. Instead, exploratory experiments and pilot studies are meant to generate new questions and hypotheses that can then be tested under more prescribed experimental conditions on a larger scale (Thabane et al, 2010).

Biological experiments are typically comparative in nature and test the “null hypothesis,” which postulates no difference between sets of biological material treated the same except for the matter in question, and that testing involves accepting or rejecting the null hypothesis with a certain level of confidence. The outcome of the experiment provides evidence about the hypothesis tested, and the amount of confidence that can be placed in the correctness of the hypothesis is dependent on the strength of that evidence and the quality of the experimentation and not on whether the results meet any arbitrary cut-off level of significance or have been published in a high-impact journal. (Fry, p. 460, reprinted with permission)

Include some examples of experiments here? Are there good examples from JAX about exploratory or pilot studies that informed subsequent experimental designs? Here’s an example of an exploratory metagenomics study that led to the identification of the proteorhodopsin gene, which led to new questions about bacterial evolution: http://philsci-archive.pitt.edu/3985/1/EE_proteorhodopsin_preprint.pdf This article has a nice table comparing theory-driven and exploratory experiments, though we may not want to go down that path. Another resource (a definition in the Encyclopedia of Systems Biology), with references: https://link.springer.com/referenceworkentry/10.1007%2F978-1-4419-9863-7_60

The remainder of this course will address the features of experimental design, including experimental units, randomization, replication, and control.

Types of controlled experiments

Pilot

(Thabane et al, 2010)

Exploratory

hypothesis generation

Confirmatory

hypothesis testing

Observational experiments

Observational studies are employed when controlled studies are not feasible. In observational experiments, assignment of treatment to subjects is outside of the control of the investigator. For example, a study of lung cancer in smokers versus non-smokers would be an observational experiment since the researcher has no control over the treatment.

Observational studies can infer causality, such as the 1956 study establishing that smoking causes lung cancer. More often, though, observational experiments suffer from confounding or reverse causation. Hidden variables or factors can confound associations between treatment and outcome. For example, a study examining the relationship between obesity and heart disease could be confounded by age, gender, or diet. The direction of cause and effect can be difficult to determine as well. For example, an association between smoking and anxiety would be difficult to interpret. Does smoking cause anxiety? Does anxiety cause smoking? Discriminating between cause and effect is challenging in observational studies.

Discussion

Identify the type of experiment in each of the following. After identifying the type of experiment, describe a next possible step. What hypotheses might the experiment generate or test? Which types of experiments have you performed?

  1. example 1
  2. example 2
  3. example 3

Solution

Discussion

If you have mainly done exploratory or pilot studies, how have or how might those studies inform experimental designs? If you have worked on experimental studies, how has that research been based on prior exploratory or pilot work? Reflect on key differences you see in study features.

Experimental units

Technical versus Biological Replicates

Challenge 1

You are attempting to get different red blood cell counts from a mouse, and you have enough blood to measure 3 times through the machine. How many replicates would this be, and of what type?
a. 3 biological replicates
b. 3 technical replicates
c. 1 biological replicate
d. 1 technical replicate
e. b and c

Solution to Challenge 1

e. One mouse means that n=1, thus 1 biological replicate. Three runs through the machine means 3 technical replicates. This would be a good way to measure variation in the instrument, but not to infer anything about mice. An n of 1 does not lead to generalizable findings.

Challenge 2

If I want to know about mean blood glucose, should I
a. measure 1 mouse 9 times
b. 3 mice 3 times each
c. 9 mice 1 time each

Solution to Challenge 2

c. 9 mice 1 time each so that you get a sense of the sample mean, the most precise estimate

Key Points

  • There are multiple kinds of biological experiments, which may involve generating hypotheses, testing the feasibility of research procedures, and testing hypotheses.

  • Pilot studies test the feasibility and efficiency of research procedures on a small scale, with few animals.

  • Experimental studies are characterized by randomization, replication, and control.

  • Exploratory studies are observational or correlational studies which identify patterns that inform hypotheses.