Introduction to research

1
Introduction to research

• HON 101 Week Two

2
What is research?

• To research is 'to investigate thoroughly' (from
  French resercher, to search)
• It is systematic process of inquiry aimed at
  discovering and interpreting facts, and the
  relationships between facts
• e.g.and especially, causal relationships between
  variables (a variable is a measurable
  characteristic or attribute of an phenomenon that
  might be expected to vary from individual to
  individual over time and space)

3
Key concepts

• Types of research
• Pure versus applied
• Primary versus secondary
• Inductive versus deductive
• Controlled versus natural experiments
• The scientific method
• Types of knowing
• Reliability versus validity
• Correlation versus causation
• Verification versus falsification

4
Pure versus applied

• Pure or basic research is the advancement of
  knowledge conducted without any particular
  practical end in mind. It provides the
  foundation for further, sometimes applied
  research
• Applied research is designed to solve practical
  problems of the modern world, rather than to
  acquire knowledge for knowledge's sake
• Examples?

5
Primary versus secondary

• In primary research data are collected
  specifically for the study at hand
• Direct observation
• Instrumentation
• Interviews
• In secondary research data is obtained from other
  sources that have collected or compiled them for
  general or other specific purposes
• Examples?

6
Secondary sources

• Published statistics
• census data, USGS etc.
• Published texts
• government reports, refereed journals and books
• Internet
• websites
• Personal documents
• diaries

7
Advantages and disadvantages of secondary data?

• Advantages
• Cheap and accessible
• Often the only resource, for example historical
  documents
• Often the only way of examine large scale and
  long-term
• Disadvantages
• Possible lack of consistency in definition of
  terms ? Biases and inaccuracies can not be
  checked ? Concern over whether data can be
  separated from the context of its collection

8
Inductive versus deductive. . .science
consists in grouping facts so that general laws
or conclusions may be drawn from them

• Charles Darwin
• What kind of reasoning is this called?

9
Kinds of scientific thinking

• Induction
• Formulation of laws based on a limited
  observations of recurring phenomenal patterns
• Deduction
• the conclusion is necessitated by, or reached
  from, previously known facts (the premises). If
  the premises are true, the conclusion must be
  true.
• the process of deriving the consequences of what
  is known
• Abduction
• inference to the best explanation, or best
  hypothesis which probably explains the data
• the process of explaining what is known.

10
Science is a way of thinking much more than it
is a body of knowledge

• Carl Sagan, astronomer
• What is this way of thinking?

11
Physics is like riding a bicycle. If you think
about how you do it, you fall off

• Richard Feynman, Physicist

12
The scientific method?

• 1. Proposition of hypotheses based on existing
  knowledge as explanations of natural phenomenon
• 2. Examination of empirical data to test
  hypotheses in order to verify hypotheses
• 3. Replication of observations or experiments to
  test reliability and validity of results
• 4. Combination of specific hypotheses into
  coherent structures of knowledge or theories
• 5. Statement of laws

13
Hypothesis

• A suggested explanation of a phenomenon or
  prediction of an outcome of observation based
  upon theorized causal relationship between
  phenomenon
• Verifcation versus falsification

14
Verification v. falsification

• Karl Popper (1902-1994) argues that science
  cannot verify only falsify
• hypothesis must be falsifiable to be scientific

15
Correlation versus causation

• Correlation
• Association or co-relation between variables.
  The departure of two variables from relationship
  of independence
• Causality always implies at least some
  relationship of dependency between the cause and
  the effect.
• Occurrence of a phenomenon B depends on the
  occurrence of a phenomenon A, where A is called
  the cause, B the effect.

16
Two assumptions of causality

• Antecedence postulates that the cause must be
  prior to, or at least simultaneous with, the
  effect.
• Contiguity postulates that cause and effect must
  be in spatial contact or connected by a chain of
  intermediate things in contact
• Note
• Due to advances in relativity and quantum
  mechanics physicists have abandoned these
  assumptions as exact statements of what happens
  at the most fundamental levels but they remain
  valid at the level of human experience

17
cum hoc ergo propter hoc

• Latin for "with this, therefore because of this
• Common logical fallacy mistaken inference when
  A happens before or at the same time as B,
  therefore A causes B
• E.g. "Those kids are always playing violent
  videogames and listening to bad music and etc...
  and they're also doing bad in school, so
  videogames and bad music and etc. must make you
  bad at school"
• Why is this wrong?
• Because both A and B can be explained by some
  external or antecedent factor C, say in the case
  of the previous example, inattentive parents.
• Or the association could be due to chance

18
An example

• Young children who sleep with the light on are
  much more likely to develop myopia in later life.
  
• University of Pennsylvania Medical Center study,
  published in Nature, May 13, 1999.
• Subsequent Ohio State Study found no direct link
  between infants sleeping with the light on and
  development of myopia
• but did find a strong link between parental
  myopia and the development of child myopia
• Conclusion myopic parents were more likely to
  leave a light on in their children's bedroom.

19
Spurious correlation

• http//www.steve.gb.com/images/science/spurious_as
  thma_transistor_correlation.png

20
Correlationcoefficients
Positive linear correlations between 1000 pairs
of numbers. The data are graphed on the lower
left and their correlation coefficients listed on
the upper right. Each square in the upper right
corresponds to its mirror-image square in the
lower left, the "mirror" being the diagonal of
the whole array. http//en.wikipedia.org/wiki/Corr
elation
21
Experiments

• Controlled experiments
• Isolation of variables in laboratory or
• drug trials etc. Comparison of control group
  with experimental group
• Natural experiments
• Cannot isolate subjects and variables.
• Problems of isolating covariance among multiple
  variables (hence use of probabalistic statistics)
  
• Unknown variables may produce spurious
  correlations

22
The scientific community

• Like all communities it has shared values
• Sharing of knowledge
• (publication and citation)
• Skepticism
• (replication and peer review)
• Objectivity
• Neutrality
• Disinterestedness

23
Homers specious reasoning

• Springfield has just spent millions of dollars
  creating a highly sophisticated "Bear Patrol" in
  response to the sighting of a single bear the
  week before.
• Homer Not a bear in sight. The "Bear Patrol" is
  working like a charm!
• Lisa That's specious reasoning, Dad.
• Homer uncomprehendingly Thanks, honey.
• The Simpsons (Season 7, "Much Apu About Nothing")
• What is Homers hypothesis?
• How does he verify this?
• Why is this specious reasoning?
• How could Homer determine the effectiveness
• of the Bear Patrol?

24
Necessary and sufficient conditions

• Necessary causes
• If x is a necessary cause of y then the presence
  of y necessarily implies that x preceded it. The
  presence of x, however, does not imply that y
  will occur.
• Sufficient causes
• If x is a sufficient cause of y, then the
  presence of x necessarily implies the presence of
  y. However, another cause z may alternatively
  cause y. Thus the presence of y does not imply
  the presence of x.

25
Ecological fallacy

• Special case of spurious correlation
• First defined by Robinson, W.S. (1950)
• "Ecological Correlations and the Behavior of
  Individuals". American Sociological Review 15
  351357.
• An example
• Correlation between blacks and illiteracy for US
  Census regions 0.91 (r2 0.83)
• Correlation between blacks and illiteracy for US
  States 0.77 (r2 0.70)
• Correlation between blacks and illiteracy at
  individual level 0.20 (r2 0.04)
• Why is this?

26
What is the ecological fallacy?

• The ecological fallacy is a widely recognized
  error in the interpretation data, whereby
  inferences about the nature of individuals are
  drawn from characteristics of aggregate
  statistics for the group to which they belong
  (e.g. correlation between group means)
• Stereotyping is an example of the ecological
  fallacy
• Popular examples?
• e.g. bad neighborhoods and crime?
• e.g. black men and crime?