Data Analysis Flashcards
What are primary sources
Information collected or observed first hand by the researcher specifically to meet the aim of their research.
What are secondary sources
Information in a study that was collected by someone else for a purpose other than the current study
Strengths of primary sources
- rich in detail
- good quality
- meets aims of study
Weaknesses of primary sources
-Time and effort - researcher must collect data before analysis can happen
- more expensive due to more time/recourses
- design, collection and analysis needs to be done by researcher
Strengths of secondary sources
- less time/ effort - no research needs to be done
- cheaper
Weaknesses of secondary sources
- data may not fit the exact aims of the study
- poor quality data = poor quality conclusions
What is qualitative data
Non-numerical - info in words that cannot be quantified eg descriptions
What is quantitive data
Numerical data - behaviour is measured in numbers or quantities e.g how much/how long
Strengths of qualitative data
-provides detailed info
- greater insight into behaviour
- draw meaningful conclusions
Weaknesses of qualitative data
- hard to draw simple conclusions/analyse
- difficult to present graphically
- bias as its open to interpretation - unscientific
What is the acronym for choosing a statistical test
I - independent groups
Really - related data
Could - correlation
Not - nominal
Calculate - chi-squared test
Sums - sign test
On - ordinal
Monday - Mann Whitney u
Without - Wilcoxon
Smiling - spearman’s rho
What is the probability used in psychology
0.05 - 5%
When may you need to use a smaller probability of 1%/0.01
Drug testing/sensitive impactful issues
If findings are significant what must you do ?
Accept the alternative hypothesis and reject the null hypothesis
If findings are non significant you must ?
Accept the null hypothesis and reject the alternative hypothesis
What are the two values you need to determine the significance
Observer and critical
All the types of graphical representation
Frequency table
Bar chart
Histogram
Line graph
Pie chart
Scatter diagram
What is normal distribution and features
Also known as bell shaped curve
Its features is the mean, median and mode are all at the exact mid point
The distribution is symmetrical around the midpoint
What is skewed distribution
Occurs where there are extreme scores in a data set
When do we get positively skewed data
There are a few extreme high scores which affects the mean
The mean is always higher than the median and mode in a positive skew
Where do we get a negative skew
If there was a very easy exam so most people get really high scores
So the mean is below the median and mode
Step by step of standard deviation
1- calculate the mean of the data
2- subtract the mean from each raw score
3- square the scores from step 2
4- add them all together
5- divide this by the total number of raw scores -1
6- square root final answer
Strengths of the mean
Most sensitive measure of central tendency because it takes into account the exact distance between all the values of data
Weaknesses of mean
Can easily be distorted by an extreme value
Can’t be used with nominal data
Does not make sense to use for small/discrete values
Strengths of median
Not affected by extreme scores so more useful
Appropriate for ordinal data so easier to calculate
Weaknesses of median
Not as sensitive as mean as the exact values aren’t represented
Strengths of mode
Unaffected by extreme values
More useful for discrete data
Only method that can be used for nominal data
Weaknesses of mode
Not useful way of describing data when there are several modes like in interval/ratio data
Level of measurement - nominal data
Categories are only named e.g hair colour or names of people
Level of measurement - ordinal data
Categories can be ordered/ranked e.g class or economic status (no clear interval between categories)
Level of measurement - interval data
Distance is meaningful e.g temperature, PH, IQ
Level of measurement - ratio
Absolute zero e.g years of work experience
Way to remember positive and negative skew graphically
Left foot = negative skew
Right foot = positive skew
Measures of dispersion - what is the range
Range is the difference between the top and bottom value of data - when you find the difference you must add 1
Strengths of range
Identifies how spread out the data is as it accounts the first and last value
Weaknesses of range
Is effected by extreme values - so may not accurately reflect
Does not take into account the distribution between numbers so we do not know if numbers are closely grouped around the mean or widely spread out
What is standard deviation
Measures the spread of the set of data, in effect the average distance of each number from the mean
Strengths of the standard deviation
Takes precise measures of dispersion allowing all values to be taken in to account - means we know whether values are closer grouped around the mean or spread out
Gives more accurate idea about how the data is distributed
Not as affected by extreme values
Weaknesses of standard deviation
Doesn’t give you the full range of the data
Hard to calculate
Time consuming
The role of the scientific community
Once a study has been carried out, a report will be written by the researcher - this report is then submitted to a research journal who may decide to publish it. The structure of these reports follow a pattern
What is the abstract
Short passage at the start of the report hat gives an overview of p’s and procedures, results and conclusion
It is a snapshot of the important info of the study sp they don’t need to read the whole thing
What is the introduction
Overview of previous research - should start B-roads then funnel to more specific to the study your conducting
Aims and hypothesis should lead on logically
Method
Detailed description of what the researcher did
Should be detailed enough that someone could replicate it to check reliability
Should include design, p’s, apparatus, procedure and ethical considerations
Results
Details about what the researcher found - can be qualitative or quantitative
Should be analysed both with descriptive and inferential statistics
Discussion
What the researcher concludes - should compare findings with previous research
Identify methodological issues and how they can be overcome
Suggest avenues for future research
References
Full titles and details of all journals/books/online texts used/referenced in the text
Very specific - title, age, page number etc
Appendix
Info that supplements the text - info that is too distracting for the main body of text
E.g a 100 question questionnaire
Acronym for the storages of a report
All - abstract
Introductions
Must - method
Reveal - results
Detailed - discussion
Relevant - references
Answers - appendix
What is the process of peer review
- scientists study something
- they then write about their results
- journal editors receives an article and sends it out for peer review
- peer reviewers read the article and provide feedback
- may accept, reject or send back with possible amendments to be then resubmitted
- when it meets editorial and peer standards it is published in a journal
What is the purpose of peer review
1- allocation of research funding ( research is paid for by govt. so reviews are required to enable them to decide which research is likely to be worthwhile
2-publication in research journals and books (prevents faulty data entering the public)
3- assessing the research rating of university departments (all unis science departments conduct research which is assessed in terms of qualifying for future funding for the department depends on the ratings at the peer review)
Peer review evaluation - strengths
Published work is validated for accuracy/validity
Peers suggest improvement and allocate further funding
Any problems/areas of weakness/ suggestions for improvement are highlighted as necessary
Upholds the principles of science - prevents scientific fraud
Adds credibility to the research and the field of study
Peer review evaluation - limitations
- takes time (up to 180 days) delays publication
- peers bring their own personal bias
- its hard to find an expert in specific or new fields
- conflict of interest
- publication bias (publishing good/successful data)
Inferential statistics - acronym
I - independent groups
Really - related data
Can - correlation
Not - nominal
Calculate - chi squared
Sums - sign test
On - ordinal
Monday - Mann Whitney u
Without - wilcoxon
Smiling - spearman’s rho
