Study sets, textbooks, questions
Upgrade to remove ads
Com 304 Midterm
Terms in this set (84)
What types of questions about communication are social-science methods best suited to address?
empirical questions, general explanations, falsifiable predictions
What are empirical questions?
Questions that can be judged, in large part, based on observations and measurements of social world,
Questions that have a "descriptive" orientation, not an explicitly "evaluative" orientation (line here is blurry)
What does it mean to "explain" communication behavior?
to relate particular events to a more general principle that helps us understand why those events occurred
What are replication and falsification, and why are they important?
replication- can be repeated
falsifiable- can be proven false if need be
What are constructs?
concept defined for specific scientific purpose
What are conceptual definitions?
it Defines a construct by relating it to other concepts
Why are conceptual definitions important?
it Clarifies what the researcher means by the construct
it Highlights the concepts "essential qualities" (to clarify what is/isn't included)
What functions do research questions play?
call for a description of something
may call for a description of a relationship between two variables
clarify what the researcher wants to know
shape/guide the entire research process
reveal the researcher's underlying assumptions
What is a variable?
a characteristic that can take on two or more values, empirical indicators of constructs
What is a Hypothesis?
declarative statements that predict a particular kind of relationship exists between two or more variables under specified conditions.
What are categorical variables?
participant are placed into different categories, but categories are not arranged from low-to-high
What are quantitative variables?
participants fall at different levels of the variable, and the levels indicate different amounts of the characteristic
What does it mean to say that levels/categories of a variable are mutually exclusive?
each unit/observation (person) falls into only one category
What does it mean to say that levels/categories of a variable are mutually exhaustive?
every unit/observation (person) falls into a category
What are the 4 levels of measurement?
What is a nominal measurement?
Numerical ranking is arbitrary
What is an ordinal measurement?
Numbers rank variable categories from high to low
What is an interval measurement?
Numbers indicate order of categories AND assume categories are equal distances apart
What is a ratio measurement?
Ratio has all an "absolute" (i.e., measurable, non-arbitrary) zero point
Which level of measurement has an absolute zero and allows us to form meaningful fractions?
What are the two types of statistics?
descriptive and inferential
What are descriptive statistics?
used to summarize responses from a sample, characterize the sample
What are inferential statistics?
used to estimate population parameters, test hypotheses about what's likely in the population (without doing a census)
What is a frequency distribution?
shows the frequency of occurrence (# and % of participants) at each category/level of ONE variable
What are the 3 key questions you should ask about a frequency distribution?
1. What is its shape? (skewness/kurtosis)
2. Where is the middle? (central tendency)
3. How spread out are the scores? (dispersion)
In a positively skewed distribution, the tail goes to the left or right?
In a negatively skewed distribution, the tail goes to the left or right?
What does it mean to say that the normal distribution is mesokurtic?
it is normally distributed
What does it mean to say that the normal distribution is leptokurtic?
it is tall and skinny
What does it mean to say that the normal distribution is platykurtic?
it is short and flat
What is the skewness indicator scale?
-1 to 1
What is the kurtosis indicator scale?
-2 to 2
What are the indicators of central tendency?
mean, median, and mode
When is it appropriate to report mode?
1. Variable is Nominal-Level (mode doesn't presume scores are ordered)
2. Distribution is Bimodal: 2 non-adjacent modes
When is it appropriate to report median?
1. interval/ratio variable, AND
2. distribution highly skewed
When is it appropriate to report mean?
1. variable is ordered
2. distribution not highly skewed and not strongly bimodal
What are three ways we can tell whether a distribution of scores is approximately normal?
1. Look at 3 indices of central tendency: how similar/different are mean/median/mode?
2. Look at histogram: how close is the shape to a symmetrical bell-shaped curve
3. Look at the skewness index (is it > +1 or -1) and the kurtosis index (is it > +2 or -2)
What is the range?
the difference between the highest and lowest scores in the distribution
What is the inter-quartile range?
the range of the middle 50% of the distribution
What is the standard deviation?
a measure of how much all scores tend to vary from the sample mean
How do you calculate the range?
How do you calculate the IQR?
75th percentile-25th percentile
How do you calculate the SD?
Calculate sample mean (X or M)
Subtract mean from each raw score (deviation score)
x = (X - X)
Square each deviation score (x2)
Sum up the squared deviation scores (Σ)
Divide by N -1 (sample size - 1)
Take the Square Root
Why is the range unstable?
Range stretched by Outliers (extreme scores)
Range is unstable: new scores = big changes
When would us the IQR?
if scores on variable are highly skewed
When would we use the SD?
if they aren't highly skewed
What two types of variables to comparison hypotheses have?
1 quantitative and 1 categorical
What two types of variables to relationship hypotheses have?
What is the difference between directional and nondirectional hypotheses?
directional actually predict which will be higher
When scores are normally distributed, what percentage of scores falls approximately within +/-1s of the mean?
When scores are normally distributed, what percentage of scores falls approximately within +/-2s of the mean?
When scores are normally distributed, what percentage of scores falls approximately within +/-3s of the mean?
What do z scores do?
specifies how far a specific unit (person) is above or below the sample mean in s units
What is the formula for z score?
When we convert raw scores to z-scores, what are the mean and standard deviation of those z-scores?
What is effect size?
indices that describe the size of a difference or an association in standardized units (rather than raw scores)
What is Cohen's d and what type of hypotheses should it be used for?
effect size index for comparison hypotheses:
What are general rules of thumb for a small, medium, and large effect using d?
d around .20 (.15-.39) is a SMALL effect
d around .50 (.40-.74) is a MEDIUM effect
d around .80 (.75-1.09) is a LARGE effect
What is the formula for Cohen's d?
What is a correlation?
degree to which a group's scores on 2 variables (X and Y) are associated: does knowing your score on X tell us anything about what your score is likely to be on Y (or vice versa)
What is a scatterplot?
graph representing the relationship between TWO (2) variables (X and Y)
What are 4 possible types of relationships between 2 variables?
1. No relationship (not related)
What is the possible range of the Pearson r?
-1 to 1
What are general rules of thumb for what counts as small, moderate, and strong relationships with Pearson's r?
r = .10 is a SMALL relationship (like d = .20)
r = .24 is a MEDIUM relationship (like d = .50)
r = .37 is a LARGE relationship (like d = .80)
What is the coefficient of determination, how do you calculate it, and what does it tell us?
When multiply r2 x 100, get the % of variance in Variable X that is shared with variance in Variable Y (vice versa)
Would we use Pearson r for a comparison or a relationships hypothesis?
What is measurement?
process of determining the existence, characteristics, size, and/or quantity of change in a variable through systematic recording and organizing of observations
What is measurement reliability?
getting consistent results when you should get consistent results
What is measurement error?
random fluctuations in scores due to factors that are temporary and shifting
How does measurement error influence measurement reliability?
as measurement error increases, reliability decreases
Why is measurement error problematic?
reduces chance of detecting systematic relationships between that measure and measures of other variables
What is temporal stability?
the degree to which people's scores on measure are stable over time
How can you measure temporal stability?
What is internal consistency?
degree to which individual survey items all tap the same thing
What are the three ways you can measure internal consistency?
1. average inter-item correlation
2. split-half correlation
3. cronbach's alpha
What are desirable levels for Cronbach's alpha?
What is measurement validity?
the degree to which a scale measures the construct it is supposed to measure
What is content validity?
the items adequately measure the construct, there is a good variety of items
What is predictive validity?
does a measure predict what it should predict? Does it predict relevant future outcomes (criterion)?
What is sampling error?
degree to which a sample statistic deviates from a population parameter
What is simple random sampling?
draw sample so that each member of population has an equal chance of being selected in sample
When is it especially important to use random sampling?
when you need a realistic estimation of the population parameter
If you choose an appropriately sized sample, what should your results look like?
the normal curve
How are sample size and sampling error related?
as sample size increases, error decreases
Sets with similar terms
Stats Exam 1
stats in psych TEST 2
Other sets by this creator
Anth 407-Exam 3
Anth 341 Midterm
Com 324 Midterm
Com 304 Final