HDFS 350 Final Exam Study Guide Flashcards
What are the major parts of a research article?
Title of project, abstract, introduction, methods, results, figures and tables, discussion, and references
What type of information is included in the abstract?
General overview of the entire study
What type of information is included in the introduction?
problem and importance
research questions
research hypothesis
research gap
past literature review
What type of information is included in the methods section?
participants: participant demographic, sample size, inclusion/exclusion criteria, recruitment and compensation
measures: description of surveys
procedures: time to finish study
stats plan: plan of statistical tests
What type of information is included in the results section?
Figures, p-value, correlation results and statistical results
What type of information is included in the discussion section?
limitations, future directions
conclusion and summary
How do you identify an independent and dependent variable in a research question?
the independent variable is the factor that is manipulated or changed by the researcher to observe its effect on the dependent variable
the independent variable is the “cause” and the dependent variable is the “effect.”
What should you include in a good difference question?
2 or more groups - If two or more groups show a significant difference in a variable/outcome
EX:Are parents and non-parents different in how much they use their cell phone while driving?
What should you include in a good associational question?
is there a significant association or relationship between two variables?
Have IV and DV, but it doesn’t matter which is which. Identify two variables instead
EX: Is there a relationship between age and time spent using a cell phone while driving?
What is a null hypothesis?
there is no difference/relationship in the data
What is an alternative hypothesis?
there is a difference/relationship in the data
How is the null hypothesis and the alternative hypothesis related?
The null hypothesis is the statement or claim being made (which we are trying to disprove) and the alternative hypothesis is the hypothesis that we are trying to prove and which is accepted if we have sufficient evidence to reject the null hypothesis
what is a p-value?
probability that the result was due to chance
what is an alpha value?
predetermined significance threshold
How are the p-value and alpha value related?
A study is statistically significant if the P value is less than the pre-specified alpha
what is nominal measurement?
name, numbers are only descriptive, categories
ex. sex, blood type, pregnancy status
what is an ordinal scale of measurement?
rank ordered categories
ex. stars of restaurant or hotel, 20 teams, order of race finish, cancer stage
what is an interval scale of measurement?
continuous; interval between values is known/equal
Differences in numbers represent real differences in the variable
ex. temperature in F or C, SAT/GRE/IQ scores
what is a ratio scale of measurement?
continuous; equal intervals; has meaningful zero (absence to the variable being measured)
e.g., length, time In seconds/minutes/hours/days, age, weight
Ratios are meaningful (twice as high; half as warm)
which scales of measurement methods are continuous?
interval and ratio
what scales if measurement methods are categorical?
nominal and ordinal
When should you create a difference question?
Only 2 groups
T-test
T-score: means of two groups/variability between groups
P-value
When should you create an association question?
3 or more groups
ANOVA
F-score: differences between groups/differences within groups
P-value
What are inferential statistics?
infer from the sample about the population
test hypothesis, draws conclusions
cannot tell us which one is correct
allow us to make inferences about the true differences in the population on the basis of the sample data
give us the probability that the difference between means (or the association) reflects random error rather than a real difference (or association)
can only tell us about probabilities in terms of our conclusions and results not certainties
what are descriptive statistics?
only describing current sample
describes sample, summary of the data
How are inferential and descriptive statistics different?
Descriptive statistics summarize the characteristics of a data set, like the mean or median, while inferential statistics use data from a sample to make predictions or inferences about a larger population
How do you determine is an effect is significant?
compare the “p-value” to a predetermined significance level (often set at 0.05);
if the p-value is less than the significance level, then the effect is considered statistically significant
What is a pilot study?
a test run of your study
why is a pilot study done
to test the feasibility of your study design
How do you determine if a distribution is normal?
visually inspect a histogram of the data to see if it resembles a bell curve (symmetrical with one peak) and compare the mean, median, and mode,
what cutoff value is used to determine if a distribution is normal?
a z-score of +/- 2
What are parametric statistics?
assume approximately normal distribution
a fixed set of parameters
what are non-parametric statistics?
do not require a normal distribution
a set of data analysis methods that make few or no assumptions about the distribution of the data being studied
What is the difference between parametric and nonparametric data?
parametric data refers to date that is assumed to follow a specific distribution where nonparametric does not make assumptions about the data
What are examples of parametric statistics?
t-test, ANOVA and person’s correlation
What are examples of nonparametric statistics?
Mann-Whitney U, Kruskal-Wallis H and Spearman Rho
What are the assumptions of parametric statistics?
the data is normally distributed or nearly normally distributed
the variances are equal across all groups
the samples are independent
the data is measured at least on an interval scale
What are the assumptions of nonparametric statistics?
the data should be obtained from a random sample representing the population of interest
each data point should be independent from the others
do not assume the data follows a specific distribution like the normal distribution
when do you use a nonparametric tests?
when the assumptions for parametric tests are not met and you are not confident that you will have normally distributed data
what are the required scales of measurement for a t-test?
interval and ratio
what are the required scales of measurement for a Mann-Whitney U test?
ordinal scale
what are the required scales of measurement for a ANOVA test?
dependent variable is measured on a continuous scale (interval or ratio)
independent variable is measured on a categorical scale (nominal or ordinal level)
what are the required scales of measurement for a Kruskal-Wallis H test?
ordinal scale
what are the required scales of measurement for a Pearson’s r correlation test?
interval or ratio scale
what are the required scales of measurement for a Spearmann rho correlation test?
ordinal scale
What are the required assumptions of a t-test?
data is randomly sampled from the population
the data is normally distributed
the data is continuous (interval or ratio scale)
the variances between groups are equal (homogeneity of variance)
What are the required assumptions of a Mann-Whitney U test?
independent samples: both groups being compared must be independently drawn from their respective populations, meaning observations within each group should not influence each other
ordinal data:the variable being measured should be at least on an ordinal scale, meaning data can be ranked from lowest to highest
What are the required assumptions of a ANOVA test?
normality (data within each group should be normally distributed)
homogeneity of variance (the variance of the data within each group should be equal)
independence (observations within each group should be independent of each other)
What are the required assumptions of a Kruskal-Wallis H test?
independence: the groups in the test must be independent of each other - meaning there is no relationship between the observation in each group or between the group’s themselves
sample size: each group must have a sample size of at least 5 observations
distribution: the data should be non-normal or have a skewed distribution
variable type: the variable of interest should be ordinal or continuous - meaning it has some kind of hierarchy
random selection: the data should be randomly selected independent samples
What are the required assumptions of a Pearson’s r correlation test?
both variables should be measured on continuous scale (interval or ratio level
the relationship between the variables should be linear
both variables should be normally distributed
there should be no significant outliers in the data
What are the required assumptions of a Spearman Rho correlation test?
the data should be measured on at least an ordinal scale
there is a monotonic relationship between the two variables
what test should be chosen if t-test assumptions are violated?
nonparametric tests such as a Mann-Whitney U test
what test should be chosen if Mann-Whitney U assumptions are violated?
a different non parametric test
what test should be chosen if ANOVA assumptions are violated?
a nonparametric test like Kruskal-Wallis test
what test should be chosen if Kruskal-Wallis H assumptions are violated?
Welch’s ANOVA
what test should be chosen if Pearson’s r correlation assumptions are violated?
Spearman’s Rho test
what test should be chosen if Spearman rho correlation assumptions are violated?
Kendall’s tau-b or Goodman and Kruskal’s gamma
When should a one sample t-test be used?
compare difference between sample mean and a certain value or number
EX: is the mean height of female college students greater than 5.5 feet?
When should you use a 2 sample or independent t-test?
compare differences between two independent groups
EX: Do parents and non-parents differ in how much they spend on clothes?
When should you use a paired sample t-test?
compare difference between two groups that are paired in someway OR compare difference before and after in one sample
EX: Do moms and their daughters differ in how much they spend on clothes?
Ex. is there a difference in stress before and after exercising
What is the APA format for reporting a mean value?
M=mean value, SD=standard deviation
ALWAYS include SD with mean
EX: Participants scored an average of 8.2 on the test (M = 8.2, SD = 1.5)
What is the APA format for reporting an r value?
r(df)=value, p=p-value
always italicize the “r” symbol
round to two decimal places
EX: There was a significant positive correlation between study hours and test scores, r(50) = .55, p < .01
What is the APA format for reporting an F value?
F(df numerator, df denominator)=value, p=value
the first number in parentheses represents the numerator degrees of freedom, the second represents the denominator
the “F” and “p” should be italicized
“F” value is typically rounded to two decimal places
EX: A significant main effect for group was found, F(2, 57) = 4.32, p = .02
How are different p-values reported?
in tables and figures: report exact p value (e.g., p=.015) unless p is <.001 (instead write as <.001)
use two decimal places for p-value greater than .01 and three decimal places for values between .01 and .001
“p” is always italicized
EX: The results showed a significant difference between groups, t(20) = 2.50, p = .02.
EX: The correlation between variables was not significant, r(50) = .15, p = .23
What does F stand for in SPSS?
indicates whether a model is statistically significant
if the Sig value is less than .05 then F is statistically significant
What is Sig in SPSS?
the p-value of the output
What is t in SPSS?
The test statistic for a t-test
higher absolute value of t indicates a larger difference between the groups being compared
lower absolute value of t indicates a smaller difference between the groups suggesting no significant difference
What is df in SPSS?
degrees of freedom which refers to the number of independent pieces of information used to calculate a statistic
sample size minus 1 (EX 20 sample size the df would be 19)
where do you look in SPSS output to know the direction of a significant difference tests?
look at the mean values listed for each group in the relevant analysis table such as the “Group statistics” section in a t-test or the “means” table in an ANOVA
what is the best visual display for correlation tests
scatter plots
what is the best visual display for differences tests
a bar chart
How is the strength of a relationship determined from a correlation coefficient?
the absolute value of a correlation coefficient
the value is closer to either +1 or -1 is a stronger relationship
a value closer to 0 indicates a weaker relationship
what is statistical power?
the likelihood of a hypothesis test detecting a true effect if there is one
how is statistical power helpful?
it helps draw accurate conclusions about a population using sample data
When is statistical power used?
to determine if a study is reasonable and ethical to conduct and to calculate the sample size needed for a study
how is statistical power related to sample size?
a larger sample size generally leads to higher statistical power
What factors determine sample size?
the desired level of confidence, margin of error, population variability (SD), effect size, power of the study and the type of analysis planned
How is effect size related to sample size?
larger effect sizes can be detected with smaller sample sizes, while smaller effect sizes require larger sample sizes
How do you determine the effect size of a correlation?
use the correlation coefficient (pearson’s r) itself as the effect size
what is a type 1 error?
rejecting the null hypothesis when it is actually true
what is a type 2 error?
when one fails to reject a null hypothesis that is actually false
what is the relationship between power and type 2 errors?
the probability of committing a type 2 error is equal to one minus the power of the test
the higher the statistical power of a test, the lower the probability of committing a type 2 error
what is statistical significance?
the probability of the null hypothesis being true compared to the acceptable level of uncertainty regarding the true answer
What is practical significance?
whether a statistically significant result from a study is large enough or meaningful enough to have real world implications
what is the difference between statistical significance and practical significance?
Statistical significance refers to whether a result from a study is unlikely to have occurred by chance, based on statistical analysis, while practical significance indicates whether the observed effect is large enough to be meaningful or relevant in a real-world context, even if it is statistically significant
what is face validity?
the degree to which a procedure, especially a psychological test or assessment, appears effective in terms of its stated aims.
What is construct validity?
the degree to which a test or measurement tool accurately captures the theoretical concept it is intended to measure
What is divergent validity?
the degree to which a test or measure is not correlated with other, theoretically unrelated constructs
What is internal validity?
the degree to which a research study can confidently establish a cause-and-effect relationship between variables
What is external validity?
the degree to which the findings of a study can be applied to other situations, people, settings, and measures
How is face validity measured?
by asking others to review the measurement technique and items, and to provide their subjective judgment
How is construct validity measured?
by assessing how well a tool measures what it is intended to measure
How is divergent validity measured?
by calculating the correlation between scores on a test designed to measure a specific construct and scores on a test designed to measure a theoretically unrelated construct
How is internal validity measured?
by evaluating how well a study design controls for extraneous variables and establishes a clear cause-and-effect relationship between the independent and dependent variables
How is external validity measured?
by considering factors like the representativeness of the study sample to the wider population
What is internal consistency reliability?
the degree to which different items on a test or questionnaire that are meant to measure the same construct produce similar results, indicating how well the items “hang together” and consistently reflect the underlying concept being measured
What is test-retest reliability?
a statistical measurement that assesses the consistency of results when the same test is given to the same group of people at different times
What is inter-rater reliability?
a statistical measure that indicates the level of agreement between two or more independent observers or raters when assessing the same phenomenon
What is parallel forms reliability?
a method for measuring the consistency of results from two different versions of a test
How is internal consistency reliability measured?
cronbach’s alpha
EX: 0.9 ≤ α: excellent
How is test-retest reliability measured?
by correlating the scores of the same test given to the same group of people on two different occasions
How is inter-rater reliability measured?
different researchers conduct the same measurement or observation on the same sample. Then you calculate the correlation between their different sets of results
How is parallel forms reliability measured?
by administering two different versions of the same test (considered “parallel forms”) to the same group of individuals, then calculating the correlation between the scores from each version
How is reliability related to validity?
Reliability and validity are concepts used to evaluate the quality of research.
They indicate how well a method, technique. or test measures something.
Reliability is about the consistency of a measure, and validity is about the accuracy of a measure
what is the reproducibility/replication crisis?
a growing concern in science where many research findings are difficult or impossible to replicate when attempted by other researchers, raising questions about the validity and reliability of those original studies, potentially undermining the credibility of scientific knowledge built upon them
What was the goal of reproducibility project?
to assess the reproducibility of scientific findings by attempting to replicate a large sample of published studies, to determine how often results could be replicated by independent researchers
What did we learn from the reproducibility project?
a significant portion of studies could not be replicated, leading to calls for improved research practices like data sharing and open science initiatives to enhance transparency and reliability of research findings
what is the file-drawer problem?
a phenomenon in research where studies with statistically significant positive results are more likely to be published, while studies with non-significant or negative results are often left unpublished and “filed away,” leading to a skewed perception of the overall research findings in a field
how does the file-drawer problem impact scientific progress?
creating a skewed view of research findings
what are the best practices for good science, both as a researcher and as a consumer?
being transparent and honest in reporting data, using rigorous methodology, properly citing sources, critically evaluating information, considering potential biases, understanding statistical methods, and actively seeking out peer-reviewed research from reliable sources
What are the features of a randomized controlled trial (RCT)?
Randomization, blinding, placebo-controlled, predefined outcomes and clinically relevant outcomes
What is a randomized controlled trial?
a research study where participants are randomly assigned to different groups, typically an intervention group receiving a new treatment or strategy, and a control group receiving either a standard treatment or no treatment
Why is the pre-test or baseline comparison between the groups important in a RCT?
it helps ensure that the groups are initially balanced on key characteristics, allowing researchers to confidently attribute any observed differences in outcomes to the intervention being studied
What should a pre-test or baseline comparison show to move forward with an RCT study?
should demonstrate that the groups assigned to different interventions (treatment and control) are largely similar in terms of key demographic and clinical characteristics
How is the effect of a randomized controlled trial statistically evaluated?
by comparing the outcomes of the treatment group (receiving the intervention) to the control group (not receiving the intervention)
what is a randomized controlled trial?
a research study where participants are randomly assigned to different groups, typically an intervention group receiving a new treatment or strategy, and a control group receiving either a standard treatment or no treatment
What do the between- and within-subject factors evaluate?
the differences between different groups of participants
What is the key way to know if a variable is a between- or within-subject factor?
see if each participant experiences only one level of the variable (between-subjects) or all levels of the variable (within-subjects)
What is the most important factor to evaluate in a RCT?
the quality of randomization and allocation concealment
how do you interpret a graph showing randomized controlled trial results results?
look for the separation between the treatment and control group lines, considering the confidence intervals to assess statistical significance, and analyze the magnitude of the effect size to determine clinical relevance
How can you identify if there is a significant interaction in a RCT test?
look for a statistically significant p-value associated with the interaction term in your statistical model
