Download Research Methods - Statistics

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

History of statistics wikipedia , lookup

Operations research wikipedia , lookup

Misuse of statistics wikipedia , lookup

Human subject research wikipedia , lookup

Transcript
Objective: 9/17/15
Today we will practice sampling & applying
concepts of statistics by completing an M&M
Lab.
Agenda:
1.
2.
3.
4.
Take out homework
Do Now-practicing IV & DV
Notes
M&M Lab
Research Methods - Statistics
★ How do psychologists ask & answer
questions? Last time we asked that we were
discussing ‘Research Methods.’
★ This time we will look at the data.
★ Stats is all about data.
★ Distinguish the difference between the purposes
of descriptive & inferential statistics.
★ Discuss the value of reliance on operational
definitions & measurements in behavioral
research.
★ Statistical procedures analyze & interpret data &
let us see what the unaided eye is missing.
★ Meaningful description of data is important in research.
★ Misrepresentation can lead to incorrect interpretation.
Descriptive Statistics
Use measures of central
tendency:
mean, median and mode
Use measures of variation:
range and standard deviation
Mean, Median, Mode, & Range
Mean, median, and mode are three kinds of "averages."
The MEAN is the “average” you're used to, where you add up all the numbers and
then divide by the number of numbers.
The MEDIAN is the “middle” value in the list of numbers. To find the median, your
numbers have to be listed in numerical order, so you may have to rewrite your list first.
The MODE is the “most” often occurring value. If no number is repeated, then
there is no mode for the list.
The RANGE is the “difference” between the largest and smallest values.
The mean is the usual average, so:
(13 + 18 + 13 + 14 + 13 + 16 + 14 + 21 + 13) ÷ 9 = 15
Note that the mean isn't a value from the original list.
This is a common result. You should not assume that your mean will be one of your original numbers.
Joe DiMaggio was the AL
MVP and had a batting avg
of .357 in 1941.
He also had an amazing 56
game hitting streak that is
still a record.
The median is the middle value, so I'll
have to rewrite the list in order:
13, 13, 13, 13, 14, 14, 16, 18, 21
There are nine numbers in the list, so
the middle one will be the
(9 + 1) ÷ 2 = 10 ÷ 2 = 5th number:
13, 13, 13, 13, 14, 14, 16, 18, 21
So the median is =
14
13, 13, 13, 13, 14, 14, 16, 18, 21
The mode is the number that is repeated more often than any other, so the mode=
13
The largest value in the list is 21, and the smallest is 13, so the range is...
21 – 13 = 8.
What is the mean?
mean = 15
median?
median = 14
mode?
mode = 13
range?
range = 8
In real life, suppose a company is considering expanding into
an area and is studying the size of containers that
competitors are offering.
Would the company be more interested in the
mean, the median, or the mode of their containers?
Answer: the mode because they want to know what size
tends to sell most often.
A Skewed Distribution
How are the results “skewed”???
Is the mean the best indicator of family income????
Probably not. That’s because of the outliers - $710,000 is one
such example above. So range & standard deviation come into
play to help be more of an accurate description of these
statistics.
Measures of Variation
Range: The difference between the highest and lowest
scores in a distribution.
Standard Deviation: Average difference between each
score and the mean.
LARGE SD: More
spread out scores are
from the mean.
SMALL SD: More
scores bunch together
around the mean.
M&M Lab for Sampling &
Statistics
1.
2.
Follow each step on the lab
Do not eat the candy until you are done!
Objective: Today we will examine
correlational studies & explore the
measures of central tendency &
measures of variation.
Agenda:
Do Now
Notes
Activity
Do Now: Turn & Talk
Abdul has volunteered to participate in an experiment evaluating the
effectiveness of aspirin. Neither he nor the experimenters know whether
the pills he takes during the experiment contain aspiring or are merely
placebos. The investigators are apparently making use of:
a.
Naturalistic observation
b.
Illusory correlation
c.
The double-blind procedure
d.
Random sampling
e.
The overconfidence effect
Explain how you came up with the correct answer.
How do you calculate Standard Deviation?
Good News - you will NEVER have to on a test.
However, I want you to understand what it represents, so here
you go!
Which of the following sets of data have the
GREATEST SD?
1, 5, 7, 30
5, 10, 12, 18
30, 32, 34, 35
How did you figure this out????
★ Can estimate SD by looking at the “spread” of #s
★ Can you find mean and compare each # to the mean
Standard Deviation
Normal Distribution: A distribution of scores that produces a
bell-shaped symmetrical curve.
In this “normal curve” - the mean, median, and mode fall
exactly at the same point.
The span of ONE SD on
either side of the mean
covers approximately
68.2% of the scores in a
normal distribution.
➔ Average IQ = 100
➔ Most people
(68.2%) fall into 85115 range
➔ IQ extremes are
above 130 and
below 70
Normal Curve
<---50%--->
<---50%--->
You need to write this
in your notes, copy it
down, & memorize it.
Period. There is no
other way to do it.
1 SD from the mean = 68.27%
2 SD from the mean = 95.43%
3 SD from the mean = 99.73%
4 SD from the mean = 99.994%
YouTube: Schallhorn on Standard Deviation
A Skewed Distribution
Remember this? Wouldn’t this skew the curve?
Is “mean” the best indicator of family income????
Probably not. That’s because of the outliers $710,000 is one such example above.
So range & standard deviation come into play to
help be more of an accurate description of these
statistics.
A Skewed Distribution: Negative vs. Positive
- Majority of scores above the mean. One or few extremely LOW
scores force the mean to be less than the median score.
+ Majority of scores below the mean. One or few extremely HIGH
scores force the mean to be greater than median score.
A Skewed
Distribution
How are the results
“skewed”???
Inferential Statistics: Involves estimating what is happening in a
sample population for the purpose of making decisions about that
population’s characteristics (based in probability theory).
Basically, inferential stats allow us to say:
“If it worked for this population, we can estimate it will work for the rest of the
population.”
ie - Drug Testing -- if the meds worked for the sample, we estimate they will
have the same effect on the rest of the population.
There is always a chance for error in whatever the findings may be, so the hypothesis
& results must be tested for significance.
Inferential Statistics
Statistical Significance - difference observed between 2 groups is probably NOT
due to chance. The difference instead is likely due to a real difference between
the samples.
Data is “significant” when the likelihood of a difference being due to chance is less
than 5 times out of 100.
In other words... There is a 95% chance (or greater) likelihood that any difference
seen is due to your independent variable shown numerically as p < .05
Important because if research is statistically significant it means that the results are
probably not a fluke or due to chance.
Inferential Statistics
Null Hypothesis - States that there is NO difference between 2 sets of data.
(basically the opposite of your hypothesis!)
Null Hypothesis = Opposite Hypothesis
Purpose...
Until the research SHOWS (by proving the original/alternative hypothesis) that
there is a difference, the researcher must assume that any difference present
is due to chance.
In other words, not due to statistical significance.
Inferential Statistics
Null Hypothesis Type I Error: Reject the null (choosing the original hypothesis), yet the null is actually
true.
Type II Error: Accept the null, yet the original hypothesis is actually correct.
**You don’t want to have errors! But, you could make them.
Truth About Population
Decision
Researcher
Makes
NULL TRUE
NULL FALSE**
REJECT NULL
(ACCEPT ORIGINAL)
Type I Error
Correct Decision
ACCEPT NULL
Correct Decision
Type II Error
** If NULL FALSE - then the
regular hypothesis is true!
Inferential Statistics
Null Hypothesis Example Original Hypothesis: “A bomb threat was called into
the front office, so we need to evacuate the school.”
Null Hypothesis: “There is no bomb in the school,
so we do not need to evacuate.”
Truth About Population
** If NULL FALSE - then the
regular hypothesis is true!
Decision
Researcher
Makes
REJECT NULL
(ACCEPT ORIGINAL)
ACCEPT NULL
NULL TRUE
NULL FALSE**
Type I Error - students
evacuated, yet bomb squad
does not find a bomb.
Erred on side of caution.
Correct decision.
Students evacuated, bomb
squad finds bomb & safely
removes it.
All are safe!
Correct decision.
No evacuation, no bomb.
Threat ignored & students
stay in class safe and sound.
Type II Error - Bomb threat is
ignored. Students stay in
class, bomb goes off &
students are injured.
Non-Experimental Research: Methods for
Examining Information
Descriptive Statistics
Correlation = strength of a relationship
between two variables
●
●
●
●
Positive vs. Negative Correlations = nature of
relationship
Coefficient of Correlation = strength of relationship
CORRELATION DOES NOT EQUAL
CAUSATION
Correlation Scatterplots
Establishing Cause-Effect
Relationships
*Inferring that one variable causes change
in another variables requires:
Demonstration that levels of the two
variables changed together
Demonstration that the cause preceded the
effect
Assurance that all other plausible causes
have been ruled out
●
●
●
Ethics in Research
1. Introduction to ethics in research
you must accept the responsibility to behave
ethically toward those who will be affected by
your research
ethics is the study of proper action
research ethics concerns the responsibility of
researchers to be honest and respectful to all
individuals who may be affected by their
research studies or their reports of the studies’
results
●
●
●
ethical issues must be considered at each step in
the research process
●
●
●
●
●
●
●
what measurement techniques may be used for certain
individuals
how researchers select individuals to participate in
studies
which research strategies and designs may be used with
certain populations and behaviors
how studies may be carried out with individuals
how results are reported
The basic categories of ethical responsibility
●
●
responsibility to the human and nonhuman individuals
who participate in the research study
responsibility to the discipline of science
2. Ethical issues and human
participants in research
Historical highlights of treatment of human
participants
●
●
●
●
World War II – brutal experiments performed on
prisoners in Nazi concentration camps
1947 Nuremberg trial with experimenters who
conducted those experiments
as a result of that trial – Nuremberg Code has been
established
●
●
10 guidelines for the ethical treatment of human participants in
research
Nuremberg Code laid the groundwork for the ethical standards
that are in place today for both psychological and medical
research
2. Ethical issues and human
participants in research
Historical highlights of treatment of human
participants (cont.)
●
●
additional examples of maltreatment of human
participants
●
●
in 1963 unsuspecting patients have been injected with
live cancer cells
in 1972 400 men had been left to suffer with syphilis
long after a cure was available
2. Ethical issues and human
participants in research
Historical highlights of treatment of human
participants (cont.)
●
●
Milgram obedience study (Milgram, 1963)
●
●
●
●
subjects instructed to use electric shock to punish other
individuals when they made errors in a learning task
participants were administering what appeared to be
dangerously strong and painful shocks
no real shocks were used in the study
although the participants in this study sustained no physical
harm, they suffered shame and embarrassment for having
behaved inhumanely toward their fellow human beings
2. Ethical issues and human
participants in research
American psychological association (APA)
Guidelines
●
●
●
www.apa.org/ethics/code.html
APA Ethics Code contains ten ethical standards,
and you should be completely familiar with all of
them before beginning any research with human
participants
2. Ethical issues and human
participants in research
American psychological association (APA)
Guidelines (major ethical issues)
●
●
No harm
●
the researcher is obligated to protect participants from
physical or psychological harm
●
●
●
Psychological harm – participants may feel increased
anxiety, anger, lower self-esteem especially in situations
where they feel that they have been cheated or insulted
any risk of harm must be justified
participants must be informed of any potential risks
2. Ethical issues and human
participants in research
American psychological association (APA)
Guidelines (major ethical issues)
●
●
Informed consent
●
●
human participants should be given complete information
about the research and their role in it
they should understand the information and then voluntarily
decide whether or not to participate
●
●
●
information – if not possible to provide the subject with information
about the purpose of the study we can explain to him at least
exactly what will be done
understanding – some participants may not be competent to
understand the research (e.g. children), therefore, it is necessary
to provide the information to a parent or a guardian
voluntary participation – participants decide to participate of their
own free will (no obligation because of a teacher or a professor
asked them to do so)
2. Ethical issues and human
participants in research
American psychological association (APA)
Guidelines (major ethical issues)
●
●
deception – informed consent can not be obtain before
the study
●
●
●
to obtain unbiased results researchers must sometimes use
deception because participants may adjust their own levels of
performance in an attempt to satisfy the experimenter
Passive deception (or omission) is the withholding or omitting
of information (researcher intentionally does not tell
participants some information about the study)
Active deception (or commission) is the presenting of
misinformation about the study to participants (misleading
participants about the specific purpose of the study)
2. Ethical issues and human
participants in research
American psychological association (APA)
Guidelines (major ethical issues)
●
●
guidelines for a study involving deception:
●
●
●
the deception must be justified in terms of some benefit that
outweighs the risk to the participants
the researcher can not conceal from the participants any
significant aspects of the study that is expected to cause
physical pain or severe emotional stress
the researcher must provide the participant with debriefing that
explains the true nature of the experiment, including the use
and purpose of deception after the study is completed
2. Ethical issues and human
participants in research
American psychological association (APA)
Guidelines (major ethical issues)
●
●
Confidentiality
●
●
is the practice of keeping strictly secret and private the
information or measurements obtained from an
individual during a research study
Anonymity
●
is the practice of ensuring that an individual’s name is
not directly associated with the information or
measurements obtained from that individuals (e.g. using
codes)
The Institutional Review board (IRB)
●
●
●
most human-participant research must be reviewed and
approved by a group of individuals (scientists and nonscientists) not directly affiliated with the specific research
study
the U.S. Department of Health and Human Services
(HHS) requires review of all human-participant research
conducted by government agencies and institutions
receiving government funds
IRB typically requires that researchers submit a written
research proposal that addresses each of the seven
criteria of IRB (minimization of risk to participants,
reasonable risk in relation to benefits, equitable
selection, informed consent, documentation of informed
consent, data monitoring, privacy and confidentiality)
●
●
●
Category I (exempt review) – anonymous survey on innocuous
topic
Category II (expedited review) – minimal risk to participants
Category III (full review) – special populations, deception,
intervention, invasive measurement
3. Ethical issues and nonhuman
subjects in research
the first ethical question is whether nonhuman subjects
should be used at all in behavioral research
APA guidelines for the use and treatment of nonhuman
subjects in research
●
●
●
●
www.apa.org/science/anguide.html
animals must be treated humanely, qualified individuals must
conduct research, the research must be justified and the
researcher has a responsibility to minimize discomfort or harm
institutions that conduct research with animals have an
animal research review board called the Institutional Animal
Care and Use Committee (IACUC)
●
●
Committee consists of a veterinarian, at least one scientist
experience in animal research and a one member of public with
no affiliation with the institution