Research Methods Unit 2 Flashcards
1
Q
What is a research/alternative hypothesis?
A
The belief that there is a difference due to the treatment/manipulated variable within an experiment (H1)
2
Q
What is a null hypothesis?
A
A null hypothesis is the belief that there will be no difference when the treatment or manipulated variable is applied during an experiment.
3
Q
What is confidence interval testing?
A
Confidence interval testing is an alternative to null hypothesis based on how certain the researcher is that their hypothesis is correct
4
Q
What is a type ll error?
A
A type ll error (the less egrigious of the two) is the incorrect belief that the null hypothesis is false and that there is not a difference when the manipulation is applied in an experiment (less egrigious because it just means the experiment failed to notice a correlation and is ammendable by replication)
5
Q
What is a type l error?
A
A type l error is the false assumption that the null hypothesis is false when it is actually true (falsely assuming there is a correlation where there is not could be the root of the replication problem.)
6
Q
What is an example of a null hypothesis?
A
When looking at normal and abnormal faces the time starting will be no differnt
7
Q
What is an example of a research hypothesis?
A
Between looking at normal and abnormal faces the participant will spend more time looking at the abnormal face.
8
Q
What is a type l error example?
A
There is a significant difference between the lookings time (where there is in fact no significant diffrence)
9
Q
What is a type ll error example?
A
There is no significant difference between the looking times (when there is in fact a significant difference)
10
Q
Univariate Analysis
A
The analysis of one variable across mutiple cases that make up a data set
11
Q
Bivariate analysis
A
The analysis of two variables across mutiple cases which make up a data set
12
Q
Multivariate analysis
A
The analysis of three or more variables across mutiple cases that make up a data set
13
Q
What are the three characteristics of a variable?
A
- The distribution of the variable
- The measure of center
- The spread/disposition of the set
14
Q
Central tendency
A
The estimate of the center of distribution of values
15
Q
Mean
A
The average of all the numbers within a distribution (is flawed because in including all the values it tends to skew high/low depending on the outliers)
16
Q
Median
A
The middle value of a distribution (the most resistant to skewing bias becaause it is a location within the set rather than a sum of values)
17
Q
Mode
A
The most often occuring values within a set (there can be multiple if a bimodal or multimodal graph)
18
Q
What are the steps to find the mean?
A
❶ Add up all the scores. That is, figure E X.
❷ Divide this sum by the number of scores. That is, divide g X by N.
19
Q
What is the means equation?
A
M = E X/N
20
Q
What are the steps to find the median
A
❶ Line up all the scores from lowest to highest.
❷ Figure how many scores there are to the middle score
❸ Count up to the middle score or scores. If you have one middle score, this
is the median. If you have two middle scores, the median is the average (the
mean) of these two scores
21
Q
How to find the mode?
A
- line up values from low to high
- count how many times each value appears
- (if there are multiple numbers that occur the same amount you have multiple modes)
22
Q
Dispursion
A
The spread of values around the central tendency
23
Q
Range
A
Subtracting the lowest from the highest value of a data set to get the range (limited due to only considering two variables and not the frequency of occurance +paired w/ the median)
24
Q
Variance
A
A measure of how spread out a
set of scores are; average of the squared
deviations from the mean.
25
How to find the variance
❶ Subtract the mean from each score. This gives each score’s deviation score,
which is how far away the score is from the mean.
❷ Square each of these deviation scores (multiply each by itself). This gives
each score’s squared deviation score.
❸ Add up the squared deviation scores. This total is called the sum of squared
deviations.
❹ Divide the sum of squared deviations by the number of scores. This gives
the average (the mean) of the squared deviations, called the variance.
26
How to find standard deviation
❶ Figure the variance.
❷ Take the square root. The standard deviation is the positive square root of the
variance. (Any number has both a positive and a negative square root. For ex-
ample, the square root of 9 is both + 3 and -3. But for the standard deviation
we always use the positive square root.)
27
What is standard deviation?
The average amount a score varies from the mean
(in a data set there should be an equal number of positive and negative values)
28
What is the standard deviation equation?
SD= (square root) E(X-M)2/ N
29
What is the sum of all sqaured?
The added value of all the sqaured scores (variance)
30
What is a normal distribution?
A normal distribution is a plotted distribution that does not change it's shape when scaled ( ex: age is a normal distribution because no matter how you change the measure of time aged it will remain the same)
31
What are the percentiles of a normal distribution?
The percentiles of a normal distribution are 34%, 14% and 2% (which is mirrored down the center of the distribution)
32
What is a z score?
A z score is a standardized unit of measure to determine where a raw score lands within the normalized graphs percentiles based on how many standard deviations it is away from the mean ( also called standard deviation scores, and only work like this when the z scores are normalized and therfore the data must fit a normalized curve)
33
How to find the z-score
1. sutract the raw score from the mean
2. divide by the standard deviation
34
What is the z-score equation
(X-M)/SD
35
What is a general experiment?
A systematic research study in which the researcher directly varies a factor holding all else constant and observe the results of systematic variation
36
What is research
Research is based on that which has come before (look to the literature to replicate)
37
What is a factor?
Another word for a variable (look to factoral analysis)
38
What is the first criteria for a true experiment?
Must have at least two groupings (usually the control and the experimental)
- in a true experiment it must be a control group
39
What is the second criteria for a true experiment?
The manipulation of a variable before the observation of the change in another variable (the independent variable must be in effect before observing the dependent variable)
(one variable has two levels/conditions the experimental and control )
40
What is the third criteria of a true experiment?
There must be random assignment to the two groups
41
What is random assignment
A condition in which an experiments participants are equally likely to end up in the experiment or control condition
42
What is random selection
Participants randomly selected by chance from a larger group
43
What is quasi-experimentation
Does not rely on random assignment (often field research) where the participants are just who are there (ex: middlemist study)
44
Independent variable
The variable of intrest to the experimentor (what they are using to manipulate the experimental condition and what is lacking in the control)
45
What is a field experiment
A controlled experiment taking place in real world conditions
46
What is field research
Research observing people in their natural environments rather than in a lab
47
What is a situational variable
A variable where so aspect of the environment is encountered
48
What is an direct intervention in the context of a situational variable
Where the participant directly interacts with the confederate or the manipulation in some way as a form of observation
49
What is an indirect intervention in the context of a situational variable?
An indirect intervention is when the participant interacts with the manipulation of the experiment in the general proximity of the manipulation
50
What is a task variable
A task variable is when the two or more groups are asked to do diffrent tasks ( ex: mazes of diffrent difficulties)
51
What is an instructional variable?
An instructional variable is when there is a diffrent in the way in which a paricpant does a task ( ex: they both are asked to memorize the same list of words but one is told about a secondary task while the other is not)
52
What is the treatment group?
The group exposed to the manipulated condition
53
What is the control group?
The group deprived of the manipulation
54
What is the comparison group
A comparative group that is given a lesser version of the maipulation than the experimental group (a factor of non-true experiments often in research replication)
55
What is an extreneous variable?
An extreneous variable is any number of uncontrolled variables that are not of intrest to the researcher (and are attempted to be controlled against but can never control against them all)
56
What is a confound?
A confound is a more egregious extreneous variable which directly interfers with the results of the data set (making it unknown what is the cause of the results)
57
What is a dependent variable?
A dependent variable is the outcome (basically the experimetors manipulate the independent variable and the depenedent variable is the resulting effect of said manipulation)
58
What is a subject variable?
A subject variable is a prexisting characteristic of a participant (ex: sex, age, physical or mental condition etc)
- these variables can not be manipulated
59
What is a manipulated variable?
An independent variable which can be manipulted by the experimentors as part of their experiment
60
Field independent
A manipulation of the envrionment to impact a specfic object
61
Field dependent
Manipulating an object within the environment to impact the overall environment
62
What is Bandura's bobo study?
This study looks at children (a subject variable) and their reaction to agression .
They do this by showing the kids different videos of real people and cartoons hitting a bobo doll before frustating them slightly and then releasing them on the toy to see how they would react (in this agression is the dependent variable and the independent variable is expposure to violence?
63
Statistical conclusion validity
Does an experiment use statistics properly ( start simplistic and increase complexity through replication)
64
Construct validity
Is the definition of the independent and dependent variable appropriate ( do they correlate well? To change this you must change the opperational definition or even what is being compared)
65
External validity
How well do the experimental findings generalize? ( impacted by the populations, environments and time period the experiment takes place in)
66
Internal validity
Th extent to which the change to the dependent variable impacts the independent variable (are they corellated to each other)
67
Pre/post test concerns
Maturation= due to the participants increase/ decrease maturity as they age
history= based on how the experiment ages over time
regression= based on the day by day performance of the participant
(all potentially impact the participants reactions before or after the experiment)
68
Selection effect
comparing two unrealted groups of participants (ex humans and rats, or college sophmores and senior citizens)
69
Attrition
Rate at which participants drop out of an experiment ( important to include the drop out numbers in data for clearest picture)
70
Between subject design
(Between group descision)
Comparing between the two levels of the independent variable, between two different groups of people.
71
Between subject factor
Between subject design of a factoral design
72
What is the first reason to use between subject design ?
Using a subject variable as the independent variable ( because it is comapring different factors of diffrent people)
(ex: 12 vs 14 year olds )
73
What is the second reason to use between subject design
When the participants in one level of the independent variable can not be exposed to the other level ( ex: drug test cant be exposed to the 2 and 4 mg tests)
74
What is the first disadvantage of between subject design ?
Large number of participants required (need equal number in each group)
75
What is the second disadavantage of between subject design
The groups need to be balanced
76
Repeated group design/ Within subject design
A smaller group of particpants are tested in both levels of the independent variable (they are their own control )
76
Random assignment
Every expeirment participant has an equal chance of being in the control or experiment group
77
Random selection
How experimentors gather their participants, by choosing a categorization and then taking a random sample from it
77
Matching
Used in small subject pool experiments to match participants together
78
Matching example
Matching 1 synethestic patient with 10 non-synethetic patients.
(other types include similar subject variables,time of test, degree of memory).
79
1st justification of matching
The matching variable will have a predictable effect on the outcome of the study
80
What are the second justification for matching
Has some reasonable way of assembling matching variables (has the potential to expose the root of the experiment)
81
Examples of relaibility and validity
- Phernology might me methologically reliable but is not a valid study
- The sat is reliable therefore it must be valid
(The idea that there something reliable must be valid but validity does not equate to relaibility)
82
1st reason for for within subject design
Need for fewer participants (less than between subject design)
83
2nd reason for within subject design
Reliant on breif exposure to the stimuli to not break the illusion of the experiment
84
3rd reason for within subject design
No need for group eqivalence because each individual is their own control group
85
4th reason for within subject design
Statistically offers a more subtle analysis than between subject design
86
What is the first disadvantage of within subject design
Prone to sequence/order effect where being exposed to the level of the independent variable and if they are exposed to it later could postively or negatively effect their results
86
87
Progressive sequence effect
When post first exposure to the independent variable they get progressively better at the task due to over exposure
( ex: 2nd trial is performed better than the 1st practice effect)
88
Fatigue sequence order effect
When after the first trial the participent gets worse at the task due to over exposure
89
Carryover effect
A change in performance occurs as a result of completing a particular sequence of conditions rather than another sequence
(ex: condition A i1st feeling cofrim condition B 2nd. Postive a transfers to B.
90
Counterbalencing
The act of controlling measures in an experiment by changing the order of stimuli
91
Complete counterbalence
A counterbalanece used when the participant is only being tested one time.
92
Partial counterbalence
Changing up the counter balences as the test goes on (from reversal, random etc)
93
Double blinding
both the experimentor and participant are in the dark about exact set of the test being administered ( more common in large group experiments)
94
Hawthorne effect
You behave differently while being watched ( has the potnetial to skew data because of it)
95
Demand characteristics
cues that could suggest the experiments goal to the participant
96
Manipulation check
the act of a distraction task or some sort of active engagement in a task to draw the participants attention to something or make sure they are activly giving accurate data
97
block randomization,
where the basic rule is that every
condition must occur once before any condition can be repeated.
98
What is an independent group single factor design?
1. A between subject design
2. Has a manipulated variable
3. Groups must be equivalent
3. Relies on random assignment (but can have matching done before the random assignment)
4. Has one manipulated variable
99
What is a Matching group single factor design
1. Between subject design
2. Manipulated variable
3. Groups must be equivalent
3. The paricipants must be matched
4. Has a single independent variable
100
What is a ex post facto/ non equivalent group single factor design ?
1. Is a between subject design
2. Non-manipulated/ subject variable
3. Groups must be unequal
4. Have the potential to be matching or non-matching
5. Must have a single independent variable
101
What is a repetative measure single factor design
1. Within subject design
2. Has a manipulated variable
3a. If tested once per trial= complete/ partial counterbalence
3b. If tested multiple times= block/ reverse counterbalanece
4. must have a single manipulated variable
102
What is an example of a independent group design
- Kasser+ Sheldon study
- Intoduced insecurity independent variable
- Group 1 influenced by insecurity and asked to write about death + group 2 to write about music (showing different tasks for those within the different groups)
- However the meat of the experiment was in the projection task afterwards
103
What is an example of matching group design
- Kroger autism study
- Matching used to group autstic participants with there general range of behavior (with the age range and neurodivergence being the commonality
- Due to small participant group it was better for the participants to be double blind
104
What is an example of ex post facto/ non eqivalent group design
- Brain damange study
- Groups matched on age, education and gender (difference= brain damage vs no brain damage)
- Pre-matching= similar but not same groups (distinction between matching test + use of subject variables_
105
What is an example of a repeated measure design ?
- same group of participants in each level
- stroop example
- read words in black in spelling a color, vs 2nd test spelling colored words in other colors ( applicable to one group for using their initial scares as a base line.
- they where also reverse counterbalanced
