Data analysis and reporting on investigations

Get Started. It's Free
or sign up with your email address
Rocket clouds
Data analysis and reporting on investigations by Mind Map: Data analysis and reporting on investigations

1. Statistical significance

1.1. Studies are designed to test a hypothesis

1.1.1. Null Hypothesis States there will be no difference or relationships between the two variables

1.1.2. Experimental Hypothesis States that there will be a difference/effect Can be directional (one condition will have a greater effect than the other) or non-directional (there will be a difference)

1.2. After the study is conducted, we need to analyse the results to calculate the probability of getting a pattern of results if they did not exist in the population

1.2.1. If this probability is small enough (p = 0.05), the null hypothesis can be rejected, if p is above 0.05, the results are not signifiant

1.3. This level of significance is designed to balance the risk of type 1 and type 2 errors

1.3.1. Type 1 error Occurs when the null hypothesis is rejected but it should have been attained Using the level of significance, there is a one in twenty chance of making a type 1 error These can be discard when the results of studies cannot be replicated

1.3.2. Type 2 error When the null hypothesis is attained but it is false

2. Summarising data

2.1. Measures of central tendency - can be used to compare two sets of data

2.1.1. Mean

2.1.2. Median

2.1.3. Mode

2.2. Measures of dispersion (spread of scores)

2.2.1. Range

2.2.2. Standard Deviation

2.3. Graphical interpretations

2.3.1. Histogram All bars are joined The width of all bars are equal

2.3.2. Scatter Graph Positive correlation Correlation coefficient: 0 to +1 Negative Correlation Correlation coefficient: -1 to 0

3. Choosing a statistical test

3.1. Is it ORDINAL data (Quantities that have natural ordering/ranking but you cannot know if intervals are equal, i.e. order of runners finishing a race) or INTERVAL/RATIO data (like ordinal data but with equal intervals i.e. seconds, meters) ?

3.1.1. Independent measures? Mann- whitney U For any N1 and N2 Observed value of U must be EQUAL TO OR LESS THAN the critical value

3.1.2. Repeated measures? Wilcoxon Observed value of T must be EQUAL TO OR LESS THAN the critical value

3.2. Is it NOMINAL data (labels with no sense of order i.e. male or female) ?

3.2.1. Chi-squared Observed value of x2 must be EQUAL TO OR GREATER THAN the critical value Also uses independent measures design

3.3. Is the data correlational?

3.3.1. Spearman's Rho Observed value of rho must be EQUAL TO OR GREATER THAN the critical value This will also use interval/ratio data

4. Dealing with qualitative data

4.1. It involves an in depth analysis of people's experiences, beliefs, attitudes etc - it is about learning from the unique experiences that people have and developing insights from this

4.2. Gathering data

4.2.1. Semi-structured interview using open ended questions

4.2.2. Participant observation

4.2.3. Focus group discussion

4.2.4. Case study

4.3. Qualitative samples tend to be a small, clearly defined group

4.4. Analysing data

4.4.1. Organising the data e.g. preparing a transcript

4.4.2. Getting to know the data thoroughly e.g. reading the transcript several times before analysis is attempted

4.4.3. Code the data Depends upon the type of analysis Interpretive phenomenological analysis Grounded theory Discourse analysis

4.4.4. Theory develops from the data during the analysis and the researchers avoid having prior assumptions whereas quantitative research is analysed in terms of existing research

4.4.5. Reflexivity When the researcher reflects on how the research activity and the researcher shape the outcome Not done in quantitative research

4.5. Evaluating data

4.5.1. Tends to be evaluated in terms of reliability and validity These concepts are difficult to apply to qualitative data as it is based on the researchers interpretation of the Ps subjective experience Tends to have high internal validity but low external validity and reliability These concepts do not make sense in qualitative research so should not be used ?

4.5.2. External Audit Involves a check of the documentation, from transcript to final analysis, by an external party This documentation should include notes about how any decisions or choices were made

4.5.3. Transferability Can the insights from the research be transferred to help understand similar situations or experience

4.5.4. Negative case analysis Exploring cases that do not fit emerging concepts

5. Reporting psychological investigations

5.1. Title

5.1.1. Should be concise and informative

5.2. Abstract

5.2.1. Brief (150-200 word) summary of the report including the synopsis of the research question, the method used, the findings and the conclusion

5.3. Introduction

5.3.1. Introduces the background/context of the study and reasoning behind it (e.g. previous research or theories), then the predictions from the hypothesis

5.3.2. Qualitative research will not have a specific hypothesis but does have aims

5.4. Method

5.4.1. How was it conducted?

5.4.2. Should have enough info for the study to be replicated

5.4.3. Includes subsections on design (and controls used), participants (and assigned conditions), apparatus, and/or materials and procedure

5.5. Results

5.5.1. Descriptive satistics Measures of central tendency and dispersion in clear tables and graphs

5.5.2. Inferential statistics Used to analyse data

5.5.3. Following this the null hypothesis can be accepted or rejected

5.5.4. Qualitative research will typically report on the analytic themes using supporting quotations

5.6. Discussion

5.6.1. How do we explain the findings? What do we learn from them?

5.6.2. Reflects the progress of scientific knowledge

5.7. References

5.7.1. Cited evidence to give people reading the repot an evidence trail

5.8. Appendices

5.8.1. Can be used for detailed information not in the report