0% found this document useful (0 votes)
8 views25 pages

Psychology Research Methods Explained

Uploaded by

kellyyycheungg42
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
8 views25 pages

Psychology Research Methods Explained

Uploaded by

kellyyycheungg42
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Chapter 2: Methods in

Psychology
PSYC& 100
Adam Wright
Basics
Science "supports" - it NEVER "proves" ideas

Empiricism: belief in an objective reality that


can be observed and known

Theory: explains a phenomenon and enables


predictions

Hypothesis: a testable guess of what will


happen
3 Types of Research (+"Mixed Methods")
Qualitative Quantitative
 Is holistic, captures the  Psychology research is
quality of an experience mostly quantitative today
 Data is words –  Data is numbers –
explanations, interviews, quantities
notes  Focuses narrowly on
 Good for topics with fine defined variables
lines and in-depth  Used to test for
examinations like case causation and
studies and correlations
ethnographies
 Used to study new
Note: It is incorrect to
areas/ideas – helps
say, "Quant. is objective
identify variables so
and qual. is subjective,"
quantitative studies can
but that false
be done later
Quantitative Relationships:
Causation
 IV - Independent Variable: the cause
 Also called the treatment variable
 Researchers change this variable and check to see what happens .
..
 DV - Dependent Variable: the (possible) effect
 After manipulating (changing) the IV, researchers measure the DV
to see what happened
 When there is a cause/effect relationship, changing the IV results
in the DV changing too

 Example: How many doughnuts can I eat at once? On different,


non-consecutive days, I ate a number of doughnuts: 2, 4, and 6
 IV = # of doughnuts eaten
 I was fine after eating 2 and 4, but got sick after eating 6
 DV = status of my digestive system
 Conclusion: I can safely eat 4 doughnuts at once, but eating 6
will make me sick
Quantitative Relationships:
Correlation Part 1
Correlation (r): when one variable is present (or
absent), is another variable also present (or
absent)? If yes, there is a correlation; there is some
kind of relationship between the variables
 A correlation does NOT mean there is a cause/effect
relationship
 Correlation research sometimes uses the terms IV
and DV, but that’s just to identify which variable the
researchers are changing or for math reasons

There is a positive correlation between sunshine


and smiles
There is a negative correlation between traffic and
Quantitative Relationships:
Correlation Part 2
Perfectly positive (r = 1): when one variable is
present, the other variable is always present
Perfectly negative (r = -1): when one variable is
present, the other variable is always absent
No relationship (r = 0): the presence or absence of one
variable doesn’t tell us anything about the other
variable

r ≈ ± .25 is a weak correlation


r ≈ ± .50 is a moderate correlation
r ≈ ± .75 is a strong correlation

r2 * 100 = the % of one variable "explained by" (not


caused by) the other variable’s presence (for +r) or
Third Variable Problem
The fact that a relationship between two properties
(variables) could be caused by a third variable (a
confounding variable), not by one property acting
directly upon the second property
 Eating ice cream and drowning have a positive
correlation. Does one cause the other?
Keeps us from knowing if a correlational relationship is
causational. I.e., it's why "correlation is not causation"
The more complex a system is, the more difficult it is to
show a causational relationship between two variables
within it

Not the same thing as an illusory correlation, which


is when a person thinks a correlation exists, but it
doesn't
Significant Results
Health news says things like, "People who eat cheese
are significantly happier, so eat cheese." BUT
"significant" in science does not tell you about the size
of a correlation. It could be r = .02! Teeny tiny!

Significance-- written as p, e.g., "p < .05"-- is a number


that represents how likely a result is to be real
"p < .05" means there is a less than 5% chance to get
this result from this sample even if there is no
relationship between the variables
 It's not quite correct to say, "there is a 5% chance that the
results of the study are due to chance alone"

When p is too large, usually when p > .05, researchers


usually do not trust that the results are real
Math Review Part 1
Mode: the most common measurement in a
study
In {3, 4, 4, 4, 7, 13}, the mode is 4

Mean: average
In {2, 4, 5, 11}, the mean is (2+4+5+11)/4
= 5.5

Median: the # (maybe not a measurement)


in the middle, with half of the measurements
above it and half below it
In {1, 1, 4, 5, 11, 13}, the median is 4.5
Math Review Part 2
Range: the numeric distance between the lowest
measurement and the highest measurement
 Range: In {3, 4, 4, 4, 7, 13}, the range is 10

Standard deviation: a number indicating how


spread apart the measurements are from the mean
 For example, "The happiness mean of the two groups,
people who play video games (SD = .7) and people
who plant gardens (SD = 1.6) is the same, so the
groups are equally happy." But wait. Are those
groups the same? The scores of the gardeners are
much more spread out. Ask yourself why might that
be and if it matters (it doesn't always).
That is a fictional example to illustrate the point
Important Terms
Validity: are they measuring what they are trying to
measure? Are they using the right tool? Must have
reliability to have validity
 Internal validity: validity of the study within itself – how
well the study does what it is trying to do
 External validity: generalizability. Results can be said
to be true of those in the population outside the study

Reliability: being able to measure twice and get same


result (if property is unchanged). Are they using the
tool correctly? Can be reliable without being valid

Bias: something that interferes with your


measurements of whatever you’re trying to measure.
There are many types, from poor methods to
researcher influences
Sampling
The method used to select participants for a
study. There are many types and methods of
sampling. The method chosen is affected by the
study’s purpose, topic, location, funding, etc.
 A group of subjects (participants) who participate
in a study is called a sample. The sample is often
divided into smaller groups, such as control
groups and treatment groups
E.g., the students attending the college who participate
in a study
 The entire group of people a study's sample will
represent is called the population. The sample is
drawn from the population
E.g., everyone attending a specific college
Sampling Methods
Convenience sample: participants chosen because they
are easy to access. E.g., studies done at colleges usually
use students as participants because students are
convenient

Purposive sample: choosing certain participants for a


study because of who they are. E.g., a study about living
through a natural disaster would want to study a group of
survivors

Random sample/selection: participants chosen using a


random system that gives every member of the population
an equal chance of being selected for the study
 Good for external validity, but difficult and expensive unless
the study simply involves a survey
 Talking to "random" people on the street is not random!
 Not used often in research, but needed for good surveys
Experiments
True "experiments" test for cause and effect. To be a true
experiment, a study must use these two things:

I. Control group: a group of participants that is observed but


not treated (manipulated) by the researcher. Control group
observations are compared against treatment group
observations, to see if and how strongly participants in the
treatment group(s) are affected by the treatments

II. Random assignment: a truly random system to decide which


participants join the control group and which join the
treatment group(s). Approximating randomness (standing in a
line and counting off, etc.) does not count as random
assignment
 Random assignment is critical because it makes groups statistically
equal across all variables. Each group will have a few of “those”
types of people (calm people, funny people, Supernatural fans, etc.)
 Note: random assignment is NOT random selection! Don't confuse
those terms!
Conducting Research
Studies must be planned out in detail before
beginning. These plans are then submitted to
IRBs, Institutional Review Boards, for approval.
Most studies using humans and many studies
using animals must get IRB approval in advance
 The primary goal of IRBs is to protect participants
and prevent unethical studies from being performed
 There are multiple codes for research ethics, along
with federal guidelines that may need to be followed
 Link to APA’s Code of Ethics (the ethics about
conducting studies are in Section 8: Research and
Publication):
[Link]
APA = American Psychological Association
3 Key APA Research Ethics
 Informed consent: Research participants must be told
what the study will be like, including potential discomfort.
Also, participants may quit the study at any time without
pressure

 Deception: Researchers can only deceive when lying is


required to obtain valid observations. By the end of the
study if not sooner, participants must be told the truth and
given the option of having any data related to themselves
discarded

 Debriefing: At the end, participants must be debriefed


and given the chance to ask questions. Participants should
leave studies feeling the same or better than when they
entered
 Deception is often explained when debriefing. It is also when
researchers ask if the participant saw through any deception and
The Process: The Scientific
Method
I. Introduction
II. Literature Review/Background
III. Methodology
IV. [Carry out the study]
V. Results/Findings
VI. Analysis/Discussion
VII. Conclusion

With little variation, this follows the order of


sections in academic research articles. Knowing
this tells you how articles are organized and how
to read them
1. Introduction
Ask a question, then find out what is known
so far
Define construct(s): define the properties
that will be studied, drawing clear boundary
lines around each property so it’s clear what
is and is not included in the study
 E.g., a person’s height, as opposed to width or
depth

May also discuss operational definitions of


the constructs, i.e., definitions of how
constructs will be measured (which test or
tool will be used)
 E.g., using a Likert-type scale to rate how
2. Literature
Review/Background
Summary of what science knows so far about
the question being asked, including how
constructs have been defined by others.
ALSO states what is NOT known (and thus,
why the study needs to be done)

Sometimes this is a separate section with a


heading, sometimes not, but it always follows
the introduction
3. Methodology
Specific details about what the researcher plans
to do, including how to select participants, what
to do to them, and how to observe/measure them

A small error in the methodology can ruin a


study’s validity. As much as possible, the
methodology must be carefully planned ahead of
time in detail. Some studies even test procedures
first to make sure they work, before doing the
“real” study

Confederate: a person who is secretly working


for the researcher. Confederates often pretend to
4. [Carry Out the Study]
Follow the methodology and conduct the
study
5. Results/Findings
Summarize and present the results that were
found. At this point, no interpretation of the
data is made. This section often includes
tables and graphs with data. It's usually not
worth trying to understand the tables before
reading the next section because many of the
numbers aren't notable and won't be
mentioned again
This section is where researchers point out
results that passed (p < .05) and results that
failed (p > .05) the significance test
6. Analysis/Discussion
Key results are re-stated and interpreted here.
This section is really important because it's
where researchers explain what they think
their findings mean

Beware of words that don’t match the


numbers: "We found a moderate correlation (r
= .15) between eating string cheese and
owning a skateboard"*
 *This is a fictional example to illustrate the point: r =
.15 isn't a moderate correlation. Don’t mistake it for
an actual research finding
7. Conclusion
Wraps up the study by offering suggestions
for further research, stating what is still
unknown, and presenting suggestions for
those studying the same or related
constructs. Some of the study's errors or
limitations might be mentioned too

Often the most opinionated section of a


research article, this is where authors really
get to say what they think
How to Read Research Studies
Research studies are not essays. You don't have to
read them in order from start to finish. A suggested
order to save time:
I. Introduction - read 1st
II. Literature Review/Background - read 2nd
III. Methodology - read 5th. Once you're familiar with
scientific procedures and can evaluate them
quickly, read this section third
IV. Results/Findings - read last or never, although you
may refer to figures here when reading
Analysis/Discussion
V. Analysis/Discussion - read 3rd
VI. Conclusion - read 4th

You might also like