Week 2 + Week 3 - Research Design & Analysis + Hypothesis testing

What are the three types of empirical research methods?

Empirical research methods: gather new information in the form of data.

• Quantitative methods: numeric data, analysed statistically. e.g., experiments, surveys.

• Qualitative methods: descriptive data. e.g., interviews, focus groups, open-ended survey responses.

• Mixed methods: combination of both approaches.

What are the 5 steps in conducting empirical research?

Step 1: Observation & Literature Review

Step 2: Aim, Research Question, & Hypotheses

Step 3: Study Design & Operationalisation

Step 4: Ethical Approval & Data Collection

Step 5: Data Analysis & Conclusion

What is a null hypothesis?

Type of statistical hypothesis

H₀ - a statement of nothing - e.g there is no difference

What is an alternative hypothesis?

Type of statistical hypothesis

H₁ - a statement that something has happened, usually the same as the research question e.g there is a difference

How can we describe distributions of data sets?

• To describe shape, focus on the presence or absence of symmetry, modes and gaps or outliers.

What is Kurtosis in terms of describing distributions of data sets?

• Kurtosis describes how “peaky” or flat a distribution is.

What is the mean and how is it calculated?

Average - The sum of the scores divided by the number of scores.

What is the median?

The value in the middle of the sequence. If there are two, you need to average (find the mean) it. They are less affected by outliers than the mean is.

What is the range and how is it calculated?

The difference (or discrepancy) between the largest score and smallest score. Largest score - smallest score = range

The range is influenced (very badly) by outliers

What is the interquartile range?

Incorporates the middle 50% of the data. Here we need to find the lower (first) and upper (third) quartile (the bottom 25% and top 25% of data), so split the sorted data up into 4 equal groups:

• 145 150 165 170 170 180 195

Interquartile range = 180 – 150 = 30

• i.e. there is a 30cm interquartile range in heights among these students (example)

The interquartile range is not influenced by outliers.

What is variance and how is it calculated?

The average squared deviation (distance or discrepancy) of scores from their mean.

What is standard deviation and how is it measured for sample sizes?

The average distance between the scores and the mean.

xi = the value

x with the line on top = the mean

n = the sample size

What is the mode and when is it relevant?

The most frequent score; the only measure of central tendency applicable to categorical variables.

What is a frequency table?

A table showing the values of the variable and the frequency of each score:

• helps identify the range of scores,

• the most and least common scores,

• and any extreme scores, usually listing values from highest to lowest and only those present.

What is a histogram?

A graphical display of a distribution where the x-axis is the variable, the y-axis is frequency, and vertical bars show counts.

• for quantitative variables there is usually no gap between bars.

What is an outlier?

An extreme score much higher or lower than the rest.

• may be genuine

• an error

• due to equipment/method issues

What is the relationship between mean, median, and mode in unimodal symmetrical distributions?

They are very close to each other.

What happens to mean, median, and mode in skewed distributions?

They can differ; in skewed distributions the mean is pulled toward the longer tail.

What is the N vs. N-1 issue in variability estimates?

N is for population; N−1 is used when estimating population variability from a sample to reduce bias; most calculators/software use N−1.

What is percentile rank?

The percentage of scores in the distribution that are lower than a given score.

What is a z-score and its formula?

The number of standard deviations a score/value is above or below the mean: z = (X − M) / SD.

Indicates how far above or below the mean a raw score is, expressed in terms of the standard deviation.

What is an outlier and its relationship to z-scores?

An extreme score that is much higher or lower than the rest of the scores in the distribution. Outliers can sometimes be defined as scores with z scores less than -3.00 or greater than +3.00 (more than three standard deviations from the mean).

What are the two different types of timing in terms of the types of research designs?

• Longitudinal – Data over time

• Cross-sectional – Snapshot at one point

What are the two different types of control & manipulation in terms of the types of research designs?

• Experimental – IV is manipulated, control group used

• Non-experimental – Observation or correlation only

What are the two different types of data sources in terms of the types of research designs?

• Survey/self-report (e.g. rating aggression)

• Behavioural/observational (e.g. giving hot sauce to measure aggression)

What is the criteria for causality?

1. Covariance – Variables must be related

2. Temporal Precedence – Cause comes before effect

3. Internal Validity – Rule out other causes

Can both experimental and non-experimental designs meet the criteria for causality?

No, only experimental designs can.

Non-experimental research designs:

• can establish #1

• (if timing correct) can establish #2

• but cannot establish #3

What is the difference between sample and population, and which one are research conclusions made about?

• Population – Group you’re interested in

• Sample – Subset you collect data from

• Research conclusions are made about the population, not just the sample

What are the 4 types of data?

Quantitative - numeric

Qualitative - categorical

Discrete - e.g rank order,1st, 2nd…

Continuous - e.g weight, temperature

How can you tell if data is quantitative or qualitative?

What are the 4 levels of measurement?

Nominal: unordered, categorical e.g gender

Ordinal: ordered, categorical e.g education level

Interval: numeric scale with consistent differences between points

Ratio: numeric scale with consistent differences between points AND absolute zero

What is the difference between absolute zero in ratios and an arbitrary zero in interval data?

Absolute zero in ratio data represents a complete absence of the measured quantity.

Arbitrary zero in interval data is just a point on the scale without representing any true absence.

E.G 0 degrees Celsius does not mean an absence of temperature (interval), while 0 degrees Kelvin represents no thermal energy (ratio)

What is an independant variable?

What you manipulate or use to predict

What is a dependant variable?

What you measure (outcome)

How are independent variables used in experimental vs non experimental studies?

Experimental designs: Independent variable can CAUSE a change in the dependant variable

Non-experimental designs: Independent variable can be associated with or PREDICT the dependant variable

What is an extraneous variable?

Extraneous variable: another variable that’s not the IV or DV

What is a confounding variable?

Confounding variable: an extraneous variable that can potentially explain the relationship between the IV and DV

Example: age, reading ability, and year of school in children

• IV: age

• DV: reading ability

• Confound: year of school

When the independent variable is categorical, what does that mean and what does it determine?

Also known as the grouping variable, often has two groups/levels (e.g., junior vs. senior staff, Australia vs. USA, attending vs. not attending open day). The IV determines the study type (experimental or non-experimental).

When the dependent variable is categorical, what does that mean?

Predicts category membership, such as the chance of passing vs. failing a test, or the likelihood of depression vs. no depression. It does not have to be limited to only two categories.