RMB

This study guide covers lecture topics such as distributions, samples and populations; normal distribution; sampling methods; ethics in research; Shapiro-Wilk test; ordinal vs. interval data; sampling variability; Bessel’s correction; Standard Error of the Mean (SEM); one and two-tailed tests; Student's t-test; Apophenia; paired samples test; Type 1 and 2 errors; Levene’s and Welch's tests; Cohen’s D value; P-values, parametric and non-parametric tests; skew distributions; Wilcoxon & Mann-Whitney tests; One-way ANOVA; solicited diaries; user-generated data; and Interpretative Phenomenological Analysis (IPA).

What is a population in research?

The total set of everyone within a group that we want to test (e.g., all primary school children in the UK).

What is a sample in research?

A subset taken from a population because we can’t test everyone.

What does a histogram represent?

The sample distribution of a value of interest.

How do bigger samples affect approximations of the underlying population?

They tend to give more accurate approximations.

What are the different sampling methods?

Random, Systematic, Opportunity/Convenience, Stratified, Cluster.

What is random sampling?

Participants are selected at random from a list.

What is systematic sampling?

A structured approach to selecting participants, where every nth participant is selected from a list.

What is opportunity/convenience sampling?

Recruitment from people closest and/or most accessible to the experimenter.

What is stratified sampling?

Recruitment aims to match key characteristics of the target population.

What is cluster sampling?

Whole groups are recruited at once, which can be combined with other methods.

Why is careful planning important during the sampling process?

It can approximate target populations; samples contain randomness, and different samples may give different results; statistics estimate variability across repeated samples.

What is the risk of biased or poorly considered sampling?

It can lead to ethical concerns and reinforce inequalities and marginalize certain groups.

What characterizes a 'Normal' distribution?

A special distribution with convenient properties, summarized by the mean and standard deviation.

What percentage of observations lie within 1 standard deviation in a normal distribution?

68.2%

What percentage of observations lie within 2 standard deviations in a normal distribution?

95.4%

What percentage of observations lie within 3 standard deviations in a normal distribution?

99.6%

When is Shapiro-Wilk used?

Provides an objective test for whether data is normally distributed (higher values = more normal data; Shapiro-Wilk p indicates significance of difference from normality).

What is ordinal data?

Data that does not have an interpretable mean and standard deviation.

What is interval data?

Numeric data that inherently has a meaningful mean and standard deviation.

What is sampling variability?

Each sample only gives an estimate of the 'true' mean.

What is the mean?

The sum of all the individual data points divided by the total number of data points.

What is the standard deviation?

The square root of the sum of the squared difference between the sample mean and each individual data point divided by the total number of data points minus one.

What does Bessel's correction do?

Makes the estimated standard deviation a bit bigger to account for the bias in estimates of the population standard deviation.

What is the Standard Error of the Mean (SEM)?

The standard deviation of the sample divided by the square root of the total number of data points.

What does the Standard Error of the Mean indicate?

The precision to which our sample mean has been estimated.

What is the correlation between the standard error of the mean and the sample size?

It decreases as the sample size grows larger, indicating larger samples are more reliable!

How are 95% Confidence Intervals calculated?

95% CI = 1.96 * SEM; (x + CI) / (x - CI)

What does the t-value grow?

It grows as the difference between the observed data mean and comparison value gets bigger.

What does the t-value shrink?

It shrinks as the variance of the observed data gets bigger.

What is the paired samples test?

One sample test comparing the pairwise differences to zero. It's for one group of people contributing two data points, and you're looking for differences.

What does the Student's t-test account for?

Uncertainty in our estimate of the mean by using the standard error of the mean. It also accounts for additional uncertainty in smaller samples with its wider tails.

What is Apophenia?

The tendency to see meaningful connections between unrelated things.

What kind of hypotheses do one sample tests examine?

Relate the mean of a sample to a prespecified comparison value.

What are Type 1 errors?

Occur when the null hypothesis is true, but we reject it.

What are Type 2 errors?

Occur when the null hypothesis is false, but we accept it.

What is a one-tailed test?

A test for strictly whether data mean is either greater than or less than a comparison value.

What is a two-tailed test?

Tests that account for both possible differences (i.e., greater than or less than a comparison value).

For T-Tests, what should you report?

Mean and standard deviation of data observations, comparison level, T-value with degrees of freedom, P-value.

What is the basic setup for a between-subjects design?

Each person contributes to a single condition.

What is the basic setup for a within-subjects design?

Each person contributes to all conditions.

What are the assumptions for a two-sample t-test?

Appropriate data type, data are normally distributed, groups have equal variance.

What is Levene's test used for?

It assesses the null hypothesis that different groups of samples are from populations with equal variances.

What is Welch's t-test used for?

Uses an UNPOOLED measure of standard deviation which is valid when the groups have different variance.

What are the commonalities between paired and independent samples t-tests?

They follow the same principle; however, independent samples are used when you're comparing the means of between two independent distributions.

What is the key use of tests when testing hypotheses?

Tests are used to compare groups by combining the size of a difference in a sample/population with the precision of the estimate.

What is Cohen's D?

Cohen's D provides a 'pure' measure of the size of a difference but does not reflect confidence in the estimate.

What is the process of Null Hypothesis Significance Testing?

Hypothesis -> Test Statistic -> Null Model -> P-Value -> Decision.

What is a P-value?

The probability of observing a result at least as extreme as the one from the data.

What does the 'degrees of freedom' of the analysis depend on?

The number of observations.

To account for uncertainty in data when computing a t-test, what procedure is taken?

The data is adjusted by using the standard error of the mean.

What does an experiment with high statistical power indicate?

That a replication has a good chance of correctly detecting the effect if it is there.

What does a test statistic quantify?

How far the observed data are from what we would expect to see from the null hypothesis.

What's the difference between Parametric and Non-parametric tests?

Parametric tests use the assumption that our data is normally distributed. However, if this isn't the case, you can use Non-parametric tests.

What does it mean if a distribution is positively skewed?

In a Positively Skewed Distribution, there are extreme values to the right.

What does it mean if a distribution is negatively skewed?

In a Negatively Skewed Distribution, there are extreme values to the left.

What do Wilcoxon & Mann-Whitney tests compare?

If we cannot compare means to test our hypotheses, we can compare medians instead. This is done by tests that involve transforming the data into ranks and comparing the observed rank-sums.

What does comparing the medians measure?

The central tendency if the data isn't normally distributed.

Under what scenarios are nonparametric tests valid?

For datasets that are normal and non-normal. Also for ordinal, interval, and ratio data plus scenarios in which you want to compare medians rather than means.

What is One-way ANOVA used for?

Provides an intuitive method to test for any possible difference between multiple groups - without prespecifying what those differences are

What are factors and levels associated with ANOVAs?

A factor is a categorical variable containing the labels of a set of groups. Levels are the different groups within a factor.

What do you do if one ANOVA detects an overall difference?

Use multiple t-tests to find where that difference is post hoc.

What does ANOVA allow for?

Quantified by dividing data point differences from the square root of the mean, for all groups. It also allows to find between that the alternative hypothesis group is better modeled with their own means.

What does a large F-factor mean for ANOVAs?

Great benefit to groups having individual means.

What does a small F-factor mean for ANOVAs?

Little benefit to groups having individual means.

Are you expected to manually input or code anything for data in computer labs?

No manual input will be requested in the multiple choice questions. However, R code might include code for reading, interpreting R code that processes data, filters, and runs analyses.

What is a critical aspect of focus groups?

Efficacious discussion and debate. Effective way in which to encourage discussion and debate. Good for ‘sensitive’ topics.

What are some examples of video prompts in research?

Beverly Hills 90210, Sex, Girls, and Kiss Curls or Meera Syal films on self-harm.

What are solicited diaries and how are they used?

Diary writing within pre-defined guidelines, intended for research purposes, stringent practices including travel and nutrition.

What is user-generated content?

Data Harvesting - Using existing forums, chats, blogs, tweets etc. and analysing that text.

When focusing on user-generated content, what data aspects should be chosen?

Select the thread and focus and download and format to be tailored to answer your research question.

What is the goal when analyzing user-generated content in social science research?

Exploring interactions, linguistic patterns/choices, community & form formation, and group norms, organization, and conflict resolution.

What are some ethical conditions related to web use?

Some may have sub-forums specific to topic, there does need to be a good fit with topic and question, and always adhere to BPS ethic for web mediated research.

What's the difference between Reflexive Thematic Analysis and Problematic TA?

Centrality of researcher subjectivity and reflexivity with the focus on deliberate and well-thought-out methodological decisions that allowed for exploration.

What is Interpretative Phenomenological Analysis (IPA)?

Attempting to understand participants’ experiences from their perspective with descriptive linguistic and conceptual comments.

What should researchers do to encourage reproducibility?

Transparency in process (methods and code), open access papers.