Scientific Methods Flashcards
Problems encountered when dealing with scientific data
Confounding variables and bias
Define Inductive reasoning
The process of reasoning that a general principle is true because the all of the special cases you’ve seen so far are true.
Example of inductive
If all people you’ve ever met from a particular town have been very strange, you might then say “all the people in this town are strange”.
Define Deductive reasoning
The process of concluding that something must be true because it is special case of a general principle that is taken to be true.
Example of deductive reasoning
If you know the general principle that the sum of the angles in any triangle is always 180 degrees, and you have the particular triangle in mind, you can then conclude that the sum of the angles in your triangle is 180 degrees.
Inductive or deductive?
If you take your medicine, you’ll feel a lot better. You take your medicine, therefore, you’ll feel a lot better.
Deductive
Inductive or deductive?
Marins first three children were boys. If she has another baby, I will be a boy.
Inductive
The scientific method
Discipline of critical thinking that subjects ideas to review and independent repetition in order to reduce the level of uncertainty we may have about how the physical world works.
Observation
Collecting and being curious about data or observations and seeing a question or a problem. Observations lead to questions that need to be answered to satisfy curiosity.
Construct hypothesis
Educated guess about how things work.
Designing an experiment
Collecting data, observations the hypothesis must be properly tested in order for the experiment to be logically valid.
Examining and interpreting
Drawing a conclusion from the results
Evaluating
Evaluating the results in the context of the hypothesis
Peer-reviewed, evaluation, publication
The work can now be examined and tested by others. The hypothesis shows whether to be true or not.
Does the scientific method prove things are true?
Science does not prove things to be true, only show things are not true.
Law
If a theory is analyzed and shown to be true, then the theory becomes a law.
Theory
Experiments and studies on a particular hypothesis and failure to show the hypothesis to be ‘not true’ the hypothesis may come to be known as a theory
A changing theory
Theories evolve and can even be completely discredited in the face of new evidence.
Alfred Wegener
His first observation was that continents seemed to fit together like a jigsaw puzzle, especially Africa and South America
Sample (of a study)
The subjects of the study, the individuals or object that were used by the experimenters to collect data
Population (of a study)
The entire group
Collecting data
Data collected
To be able to determine if the appropriate data were collected, you need to understand the purpose of the study or hypothesis.
Confounding variables
Confounder is a variable that influences both the dependent variable and independent variable.
How the independent variable, to see whether that has any affect on the dependent variable.
Independent variable
=amount of ice cream
Dependent variable
= # of people who eat the ice cream
Bias
Favour or against something
Selection Bias
Choosing samples for an experiment or bias in setting the parameters for an experiment.
Funding bias
Reporting results which favour the funding organization.
Confirmation bias
The tendency to use data supporting your beliefs.
Publication/reporting bias
Bias un publicizing certain results over other results, or a bias in reporting unexpected results.
4 types of Bias
Selection bias
Funding bias
Confirmation bias
Publication/reporting bias
Types of selection bias
Sampling bias, Time interval bias, Particular bias, Data bias, Attrition bias
Sampling bias
The participants/items selected for a study may introduce a bias in your analysis
Time interval bias
Arbitrarily choosing to end an experiment when it supports a certain claim.
Participation bias
When individuals volunteer to participate in a study, three might be participation bias
Data bias
Choosing to use only a certain subject of the data
Attrition bias
Discounting trial subjects/tests that didn’t not complete a study.