Designing an Experiment
Designing an Experiment
This lesson reviews scientific reasoning and the scientific method. This introduces experimental design and the factors one must consider to build a successful experiment.
Scientific Method
Scientific Reasoning
When conducting scientific research, two types of scientific reasoning can be used to address scientific problems: inductive reasoning and deductive reasoning. Both forms of reasoning are also used to generate a hypothesis.
Inductive reasoning involves drawing a general conclusion from specific observations. This form of reasoning is referred to as the “from the bottom up” approach. Information gathered from specific observations can be used to make a general conclusion about the topic under investigation. In other words, conclusions are based on observed patterns in data.
Deductive reasoning is the logical approach of making a prediction about a general principle to draw a specific conclusion. It is recognized as the “from the top down” approach. For example, deductive reasoning is used to test a theory by collecting data that challenges the theory.
For Example
Use your inductive reasoning to determine the next item in the sequence of events:
1. fall, winter, spring . . .
2. 4, 8, 12 . . .
Did You Know?
While Francis Bacon was developing the scientific method, he advocated for the use of inductive reasoning. This is why inductive reasoning is considered to be at the heart of the scientific method.
Designing an Experiment
According to the scientific method, the following steps are followed after making an observation or asking a question:
- conduct background research on the topic
- formulate a hypothesis
- test the hypothesis with an experiment
- analyze results
- report conclusions that explain whether the results support the hypothesis
This means after using logical reasoning to formulate a hypothesis, it is time to design a way to test this hypothesis. This is where experimental design becomes a factor.
Experimental design is the process of creating a reliable experiment to test a hypothesis. It involves organizing an experiment that produces the amount of data and right type of data to answer the question. A study’s validity is directly affected by the construction and design of an experiment. This is why it is important to carefully consider the following components that are used to build an experiment:
- Independent variable: This factor does not depend on what happens in the experiment. The independent variable has values that can be changed or manipulated in an experiment. Data from the independent variable is graphed on the x-axis.
- Dependent variable: This factor depends on the independent variable. Recognized as the outcome of interest, its value cannot change. It can only be observed during an experiment. Data from the dependent variable is graphed on the y-axis.
- Treatment group: This is the group that receives treatment in an experiment. It is the item or subject in an experiment that the researcher manipulates. During an experiment, treatment is directly imposed on a group and the response is observed.
- Control group: This is a baseline measure that remains constant. Used for comparison purposes, it is the group that neither receives treatment nor is experimentally manipulated. One type of control is a placebo. This false treatment is administered to a control group to account for the placebo effect. This is a psychological effect where the brain convinces the body that a fake treatment is the real thing. Often, experimental drug studies use placebos.
Test Tip
It can be hard to remember the differences between an independent and a dependent variable. Use the following mnemonic to help keep those differences clear:
D = dependent
I = independent variable
Y = y-axis
X = x-axis
M = manipulated variable
R = responding variable
Data Analysis and Interpretation
When researchers test their hypotheses, the next step in the scientific method is to analyze the data and collect empirical evidence. Empirical evidence is acquired from observations and through experiments. It is a repeatable form of evidence that other researchers, including the researcher overseeing the study, can verify. Thus, when analyzing data, empirical evidence must be used to make valid conclusions.
For Example
Studies have shown there is a positive correlation between smoking and lung cancer development. The more you smoke, the greater your risk of developing lung cancer. An example of a negative correlation is the relationship between speed and time when distance is kept constant. The faster a car travels, the amount of time to reach the destination decreases.
While analyzing data, scientists tend to observe cause-and-effect relationships. These relationships can be quantified using correlations. Correlations measure the amount of linear association between two variables. There are three types of correlations:
Positive correlation:
As one variable increases, the other variable also increases. This is also known as a direct correlation.
Negative correlation:
As one variable increases, the other decreases. The opposite is true if one variable decreases. A negative correlation or negative variation is also known as an inverse correlation or an indirect correlation.
No correlation:
There is no connection or relationship between two variables.
From graphs to tables, there are many ways to visually display data. Typically, graphs are a powerful way to visually demonstrate the relationships between two or more variables. This is the case for correlations. A positive correlation is indicated as a positive slope in a graph, as shown above. Negative correlations are indicated as a negative slope in a graph. If there is no correlation between two variables, data points will not show a pattern.
Scientific Tools and Measurement
Researchers use a wide variety of tools to collect data. The most common types of measuring tools are outlined below:
- Barometer: Used to determines the air pressure in a space.
- Clock or stopwatch: Used to record time.
- Graduated cylinder: Used to measure the volume of liquid.
- Ruler: Used to measure the length of an object.
- Thermometer: Used to measure temperature. Measurement values may be expressed in degrees Celsius or Fahrenheit.
- Triple beam balance: Used to measure the mass of an object or to determine the unit of mass. Electronic balances are used to measure very small masses.
Measured values are often associated with scientific units. Typically, the metric system is preferred when reporting scientific results. This is because nearly all countries use the metric system. Additionally, there is a single base unit of measurement for each type of measured quantity. For example, the base unit for length cannot be the same as the base unit for mass. The following base units are used:
| Unit of Measurement | Base Unit Name | Abbreviation |
| Length | Meter | m |
| Mass | Gram | g |
| Volume | Liter | L |
Another benefit of the metric system is that units are expressed in multiples of 10. This allows a researcher to express reported values that may be very large or small. This expression is facilitated by using the following metric prefixes, which are added to the base unit name:
| Prefix | Abbreviation | Value | Description |
|---|---|---|---|
| kilo | k | 1,000 | thousand |
| hecto | h | 100 | hundred |
| deka | da | 10 | ten |
| BASE | N/A | 1 | one |
| deci | d | 0.1 | tenth |
| centi | c | 0.01 | hundredeth |
| milli | m | 0.001 | thousandth |
Let’s Review
- Formulating a hypothesis requires using either inductive or deductive reasoning.
- A good experimental design properly defines all variables and considers how data will be analyzed.
- Correlations illustrate the cause-and-effect relationships between two variables.
- Positive and negative correlations can be displayed graphically by analyzing the slope of a line.
- Different devices are used to measure objects in an experimental study.
- The metric system is usually used when expressing the units of measured values.
Scientific Method and Designing an Experiment Flashcards
Subscribe to the online course to gain access to the full lesson content.
If your not ready for a subscription yet, be sure to check out our free practice tests and sample lesson at this link