Research methods

• Observational studies 


  Behavioural categories, event sampling, and time sampling all simplify the process of observing behaviour and make it more reliable.

  But they can also simplify and reduce the level of detail, and so reduce the validity of the observation.
• Questionnaires


  Closed questions

  Questionnaires tend to make use of closed questions, where a selection of options are provided. These include:

  • Yes-no questions.
  • Multiple choice questions.
  • Likert scales, such as where a scale from 1-7 is used to indicate how strongly someone agrees or disagrees with a statement.
• Questionnaires - EVAL

  Strengths:

  • Reliable – easy to replicate.
  • Gather lots of data quickly.
  • Closed questions generate quantitative data and are easy to analyse.

  
  

  Weaknesses

  • lacks validity - close questions limit the participant's answers
  • social desirability - demand characteristics - decrease validity
  • lack of contact can make answers less valid as the participant isn’t able to explain what they mean by certain answers - follow-up like in interviews
• Variables

  Confounding variables

  • If an outside variable changes across the two conditions, this becomes a confounding variable and can ruin the experiment.
  • In the earlier example, if participants in one condition were given more time to study the facts than those in the other condition, this would be a confounding variable.
  • It would make it impossible to be sure whether the IV (memory technique) had had an effect on the DV.
• Variables

  Extraneous variables

  Some variables can’t be entirely eliminated, such as background noise and temperature.

  These are called extraneous variables.

  The researcher tries to minimise the effects of extraneous variables as far as possible.

  Any remaining extraneous variables are a source of random error, but do not invalidate the experiment.
• Counterbalancing

  To minimise order effects, experiments use a technique called counterbalancing.

  This involves allocating participants to conditions as follows:

  • Half of the participants complete condition 1 first, and condition 2 second.
  • The other half of the participants complete condition 2 first, and condition 1 second.

  Counterbalancing does not get rid of order effects, but it stops order from becoming a confounding variable because the effect is balanced out over the two conditions.
• Demand characteristics and social desirability 


  Controlling factors

  Both demand characteristics and social desirability bias can be controlled by using a single-blind technique where the participant is unaware of the true aim of the research, but this can lead to ethical issues as it involves deception.
• Limitations of ethical guidelines
• Limitations of ethical guidelines

  • complexity - ethical judgements are never black and white - do the ends (scientific breakthrough) justify the means (unethical) - however, participant's rights should always come first

  Socially sensitive research

  • Even if ethical guidelines are stringently followed, there is still the possibility that the research may be unethical as it could be socially sensitive - research where there are potential social consequences for the participants, or the groups of people represented by the research.
• Peer-review


  other scientists provide feedback on the study - allows high quality work to be published

  Peer-reviewed journal:

  • read by several other scientists (generally anonymously) with expertise in the subject matter.
  • peer reviewers provide feedback - to both the author and the journal editor -regarding the quality of the draft.
  • Ultimately, the journal editor will compile all of the peer reviewer feedback and determine whether the article will be published in its current state (a rare occurrence), published with revisions, or not accepted for publication.
• Peer-review


  Peer-review feedback

  Peer reviewers look for a strong rationale for the research being described, a clear description of how the research was conducted, and evidence that the research was conducted in an ethical manner.

  They look for flaws in the study's design, methods, and statistical analyses.

  They check that the conclusions drawn by the authors seem reasonable given the observations made during the research.

  Peer reviewers also comment on how valuable the research is in advancing the discipline’s knowledge.
• The Implications of Psychological Research for the Economy


  Psychological research has an impact on science and our understanding of human behaviour. But it can also have practical benefits to the economy.

  Researchers also want to demonstrate that their research has a positive impact on society.

  There are numerous examples of how psychological research can contribute to people’s well-being, make citizens more productive, and boost the economy by reducing absence from work.

  It can also reduce costs to society, allowing money to be invested into other things such as technical innovations.
• Implications of psych research for the economy 


  • Clinical psychology - understand and treat mental health problems, which are among the greatest causes of days lost to work worldwide.
  • Educational psychology - understand mood, motivation and effective learning - can be applied in schools to increase attainment and by employers to boost productivity.
  • Forensic psychology - understand and tackle crime - eg study of eyewitness testimony - reduces costs to society e.g. from unfair convictions, and increases the supply of potential workers.
• Peer-review 


  Process:

  1. look how and why the study was conducted
  2. look for flaws in the design, method and statistical analyses
  3. check how reasonable the conclusions drawn are
  4. comment on how valuable the research is overall
• Reliability 


  how consistent results are - if the study was repeated in identical conditions or with similar participants, come up with the same results?

  ways to assess:
• Reliability 


  how consistent a test is - if the study was repeated in identical conditions or with similar participants, come up with the same results?

  ways to assess:

  • test-retest - study is repeated on two separate occasions - similar results - high external validity
  • split-half method - when the test is split into two - the same participant does both halves - similar results - high internal validity
  • inter-observer reliability - two observers observe the same behaviour at the same time but separately, previously agreed behavioural categories - 0.8+ correlation
• Empirical/ scientific Method

  1. ask a question - observe a phenomenon or group of phenomena
  2. form a theory - A theory is made based around existing knowledge, as well as the researcher's questions
  3. make a prediction - formulate a hypothesis - precise and testable statement
  4. carry out an experiment/test/study - The experiment must have a standard procedure to follow, with controls to allow for validity and reliability - The results of the study will provide evidence for the researcher’s theory.
  5. adjust the theory - hypothesis can be accepted or rejected - If rejected, the theory would be altered
• Qualitative data can be converted into quantitative data by categorising written answers into different types, and then calculating the percentage of each type.
• Descriptive statistics

  Central tendency: mean, median, mode

  Measures of dispersion: range and standard deviation - carried out following a simple formula, and shows the typical amount by which scores differ from the mean.

  
  

  The mean and standard deviation are more powerful statistics, as they take every score into account.

  But the other statistics are also useful, especially when describing skewed sets of data.
• Correlation - Evaluation 


  Advantages

  • They measure the strength of a relationship.
  • They are a 1st step in research.
  • Disadvantages
  • They cannot show cause, only a relationship.
  • They do not reflect curvilinear relationships.
• • A histogram shows a continuous variable along the horizontal (x) axis, such as the age range of different participants, and line graphs have a similar purpose.
  • A pie chart should represent fractions of a whole. It is not the right choice for showing the DV in a study, but could be used to describe characteristics of a sample.
• • Scattergrams are only used to show correlations, not the results of experiments.
• Data distributions


  Negative skew

  In a negatively skewed distribution, the mode is a higher value than the mean and median. The skew is to the left.

  If the distribution of A level results were negatively skewed this would mean that more students were achieving the higher grades than the lower grades and the modal grade might be a B. In this case, the exam board might raise the grade boundaries in order to achieve a more normal distribution of results.
• Data distributions


  Positive skew

  • In a positively skewed distribution, the mode is a lower value than the mean and median. The skew is to the right.
  • If the distribution of A level results were positively skewed this would mean that more students were achieving the lower grades than the higher grades and the modal grade might be a D. In this case, the exam board might lower the grade boundaries in order to achieve a more normal distribution of results.
• Levels of measurement

  the difference between interval and ratio data is that interval data has an arbitary zero point whereas there is no absolute zero point in ratio data

  • eg zero degrees celcius (interval) doesnt mean there is no temperature whereas zero pounds (ratio) means no money left
• • Content analysis tries to quantify (put into numbers) qualitative data.
  • Essentially, all the detail is broken down into either categories.
  • By breaking the data down into categories, the data can be more objectively analysed.
  • Rather than looking at the entire collection of interviews/transcripts/etc., a representative sample of the qualitative data is taken.
  • The data then is analysed according to coding units.
  • Coding units are another way of saying categories.
  • It is crucial that the coding units are operationalised.
  • Operationalising improves the validity and reliability of the results.
• Thematic analysis


  • This method is used quite often for analysing qualitative data.
  • This form of analysis involves the scientist looking over their data to familiarise themselves.
  • This allows them to establish the key themes within the data - themes they can define and name.
  • From there, they can write a report.
  • Finally, hypotheses can be formulated from this data.
  • The hypotheses, even though they could still be subjective, are based on the themes of the data.
• Thematic analysis


  Strengths:

  • Unlike content analysis, the level of detail is maintained.
  • Objectivity is possible.
  • This is done through triangulation, which involves comparing other sources of data (like previous interviews) to verify the conclusions drawn.
• when calculating a sign test do you count the no change/ equal sign as a sign in the calculation/ tallying of the number of participants (N)?

  NO
• Scattergrams are only used to show correlations, not the results of experiments.
• Naturalistic observation

  One of the best ways to gain information is to observe the behaviour in its natural context. But people might change their behaviour in unexpected ways if they know they are being observed. So, for a researcher to get accurate information, a researcher must be unobtrusive and inconspicuous. This type of observational study is called naturalistic observation: observing behaviour in its natural setting.

  • main advantage: validity of the data that is collected. The findings of the research are easily generalisable and can be applied to real-world situations.
• Limitations of participant observations

  Participant observation also has its limitations - the presence of the researcher may make people suspicious or otherwise change their behaviour, meaning that the researcher does not see authentic behaviour.

  It is also unethical for observers to assume a fake identity to observe behaviour.
• Limitations of observational research

  Observational research is often used when studying individuals who are interesting to researchers because they have a rare characteristic.

  As a result, the focuses of the study are not like most other people.

  If scientists ultimately want to explain all behaviour, focusing attention on such a special group of people can make it difficult to generalise any observations to the larger population as a whole.
• observational research



  weaknesses:

  Observational research is often used when studying individuals who are interesting to researchers because they have a rare characteristic.

  If scientists ultimately want to explain all behaviour, focusing attention on such a special group of people can make it difficult to generalise any observations to the larger population as a whole.

  Case studies do provide enormous amounts of information. But since the cases are so specific, the potential to apply what’s learned to the average person may be very limited.
• Interviews


  Strengths:

  • Interviews tend to generate qualitative data, which is in depth and detailed.
  • Semi-structured interviews can lead to serendipitous findings.
  • Interviews have validity as a rapport develops between researcher and interviewee.
  • Interviewer can explain the questions.

  Weaknesses:

  • Researcher bias – this is a real issue and can be as a result of what the researcher says or simply their body language and general manner.
  • Demand characteristics & social desirability.
  • Difficult to analyse.
• Correlations


  Correlation can be statistically represented by a correlation coefficient. The strength of the correlation is shown by how far from 0 it is - the further from 0 the stronger the correlation. The direction of the correlation is shown in the sign. Positive means the variables move in the same direction, negative means different directions.