Research methods

• One limitation of correlational studies is that they cannot establish causality. It's unclear whether one variable causes another or whether both variables are caused by a third factor. Another limitation is that correlation does not imply causation - just because two variables are correlated doesn't mean one causes the other.
• Correlational studies are useful when investigating relationships between variables, such as the relationship between genetics and personality traits. Correlations can be calculated using statistical methods, allowing researchers to determine whether two variables are related. These studies are also relatively easy to conduct and require fewer resources compared to experimental studies.
• Correlational research designs are useful when studying complex behaviours like aggression because they allow researchers to examine multiple variables simultaneously. Correlation coefficients can be calculated to assess the relationship between two variables, providing insight into the extent to which one variable predicts another. For example, a correlation coefficient of +0.5 indicates a moderate positive association between two variables, while -0.7 suggests a strong negative association.
• Experimental studies are useful when investigating causal relationships between variables. By manipulating independent variables (IV) and measuring dependent variables (DV), researchers can determine whether changes in IV lead to changes in DV. This approach provides stronger evidence for causality compared to correlational studies.
• Correlational studies are useful for identifying relationships between variables and making predictions based on those relationships. They are also relatively cheap and easy to conduct, allowing researchers to collect large amounts of data quickly. However, they suffer from limitations such as confounding variables, reverse causality, and lack of control over extraneous variables.
• A strength of correlational studies is that they can identify patterns and trends in behaviour without imposing artificial conditions on participants. This allows researchers to observe naturalistic behaviour and draw conclusions about real-world situations.
• However, there are limitations to correlational studies. One major limitation is that they do not prove cause and effect. While correlational studies can show that two variables are associated with each other, they cannot demonstrate that one variable directly causes the other.
• Lab experiment

  • Manipulated Iv and measured Dv
  • controlled variables
  • artificial environment
  • allocation of participants to groups
• Field experiment

  • Manipulated Iv and measured Dv
  • controlled variables
  • natural environment
  • allocation of participants to groups
• Natural experiment

  • Natural Iv and measured DV
  • no control of variables
  • natural environment
  • no allocation of participants to groups
• Quasi experiment

  • Natural Iv and measured DV
  • Controlled variables
  • Artificial environment
  • no allocation of participants to groups
• Cross sectional study Pros

  • Quick + cost effective
  • Easier to analyse
• Longitudinal studies pros

  • Less sensitive to timing so can identify patterns
  • Causation can be established
• Cross sectional study cons

  • Static view as snapshot in time ( decreases validity because…)
  • Causation can’t be established
• Longitudinal studies cons

  • Time consuming and costly
  • Complexity in analysis
• We can use Chi squared when it is a test of difference with unrelated data (independent measures) and the data is Nominal/categorical for example hair colour or men and women.
• We use spearman’s rho for a test of relationship with ordinal data (data that can be ranked) for example happiness out of 10 while you’re in school.
• Negative skew

  
• Positive skew

  
• We use Mann Whitney for a test of difference using unrelated data (independent measures) and the data is ordinal which means it can be ranked for example scores on a maths test out of 20
• The calculated value is less/more than the critical value so the results about [context] are/aren‘t significant for a one/two tailed test. (1)

  There is a greater than 95% chance {if P<0.05} that [context]. Or any difference about [context] is due to chance and chance alone.
• When making conclusions from results always reference the data presented in the table or graph e.g 3.2 seconds
• Strength of ranked scale questions is that they provide numerical data that can be easily analysed increasing the validity.
• Weakness of ranked scale questions is that they are subjective since different participants may interpret the scale differently. Due to this many people tend to rank towards the middle which can lead to inconsistencies in the data collected decreasing the reliability.
• For example to include context

  There is not a significant difference between the choice of gender stereotypical toys by boys and girls , as the calculated value (0.24) was less than the critical value (2.71) for a one tailed test at p=0.05
• Two reasons why a pilot study may be carried out (in context)

  • to check data (about context) would be significant
  • To make sure questions asked (about context) make sense/ are clear
  • To make sure study (about context) is within ethical guidelines
• We use Wilcoxon when it is a test of difference and the data is related (repeated measures or matched pairs) The data is also ordinal so it can be ranked such as test scores.
• self reports involve asking participants about something so they can report it themselves
• One strength of self reports is that they allow researchers to investigate attitudes beliefs and opinions which can’t be easily discovered by other methods. This means results are more likely to be accurate since insight can be gathered from a perspective not possible through the use of other methods increasing validity
• Another strength of self reports is that they are easy to do, less time consuming and cost effective which increases their practicality
• One weakness of using self reports (for context) is that the data collected is often unique to individuals and groups. This means results would be less representative of those that didn’t do a self report (on context) decreasing the generalisability
• Strength of quantitative data

  Objective as Can be statistically analysed to make comparisons about (context) which increases the accuracy of the data collected about (context). Therefore increasing the validity.
• Weakness of quantitative data

  It doesn’t allow for any detailed information about (context) to to be collected and why they feel that way, so may be less accurate as it is not a true representation of the participants thoughts on (context). Decreasing the validity.
• What do experiments involve in terms of variables?

  Manipulation of variables to measure factors
• What do types of experiments depend on?

  Environment and nature of the independent variable
• How do types of experiments vary?

  By laboratory or field experiment type
• What are the characteristics of a laboratory experiment?

  • Manipulated IV and measured DV
  • Controlled variables
  • Artificial environment for tasks
  • Allocation of participants to groups
• What is a key feature of a natural experiment?

  No control of variables
• What defines a field experiment?

  Natural environment with manipulated IV
• What is a characteristic of a quasi-experiment?

  Groups are not randomly assigned