🔥 Data Handling: Types, Interpretation & Display of Data

Quantitative Data

• numbers

• can be transformed into tables, graphs, charts, percentages

How is Quantitative Data analysed

• statistically analysed

• e.g. mean, mode, range (descriptive stats)

• mann whitney, spearman’s rho, related t-test (inferential stats)

What type of Research Methods tend to generate Quantitative Data

• experiments e.g. scores on test

• observations e.g. tally charts

• correlations e.g. correlation coefficient of +0.7

• questionnaires using closed questions

Reliability strength of Quantitative Data

• tends to be reliable

• as is easy to analyse and compare

• because techniques used to collect it are replicable

• e.g. standardised, meta-analysis

Generalising strength of Quantitative Data

• highlights trends and patterns

• which is useful when researchers wish to apply general laws of behaviour

Why criticism of Quantitative Data

• can reveal what behind behaviour but not why

• lacks explanatory power

• lowers its validity

Usefulness criticism of Quantitative Data

• quantitative data tends to over-simplify

• complex, multi-faceted nature

• of human behaviour and experience

• limits usefulness as means of gaining insight into motives, dreams, fears etc

Qualitative Data

• words or images

• e.g. thoughts in a diary

• feelings etc in an interview

• painting created to express inner turmoil

• focus group interview

Qualitative Research Methods include

• interviews

• diary entries

• naturalistic observations

• open-ended questions

Ecological Validity strength of Qualitative Data

• allows researchers to gain insight into nature

• of individual experience and meaning

• makes it high in ecological validity

Understanding strength of Qualitative Data

• can be used to expand on and deepen knowledge

• of complex behaviours

• case of HM involved man with extreme memory loss

• interviews and observations of HM shed light and helped confirm quantitative results

• e.g. memory tests

Generalisability criticism of Qualitative Data

• tends to use small sample sizes

• results are difficult to generalise

• to a wider population

Subjective criticism of Qualitative Data

• qualitative methods are subjective in nature

• does not embrace features of science

• e.g. lack of objectivity and control

• lacks reliability

What is Primary Data

• collected at the source

• researcher collects two sets of scores after running an experiment

• researcher conducts a questionnaire

• from which are able to analyse range of responses

Reliability strength of Primary Data

• may be more reliable and valid than secondary data

• as researcher has full control over how data is collected

Trustworthy strength of Primary Data

• more trustworthy than secondary data

• researchers know research will be subjected to peer review

• if negative = harms reputation

• makes sense for researcher to take necessary care to present best designed and delivered study possible

Power criticism of Primary Data

• primary data is derived from single study

• compared to secondary data

• which can amass huge samples

• limits potential statistical power of primary data

Expense criticism of Primary Research

• primary research is expensive and time consuming

• compared to using secondary data

• which can be gathered quickly

• if researcher doesn’t find significant result, may feel time and money spent was wasted

What is Secondary Data

• any research findings which are pre-existing

• not been collected at source

• obtained by other researchers

• allows non-interested researcher to gain overview of topic

• multiple sources (meta-analysis)

Confidence strength of Secondary Data

• already been peer-reviewed

• e.g. meta-analysis

• significance of each study been established

• time & money not been wasted and researchers can have confidence in data

New Insight strength of Secondary Data

• may provide new insight into new theories and research

• several studies on same topic are analysed

• allows researcher to see patterns, trends or interests

• unlikely to emerge with analysis of one study

Not Direct criticism of Secondary Data

• may not directly address aim of topic of research

• researchers lack of familiarity with data

• means they misinterpret some aspects of og research

• affects validity of secondary data

Control criticism of Secondary Data

• researcher has not run og studies themselves

• do not know degree of control and rigour

• exercised by original researcher

• lack of control affects reliability of data

What are tables used to present

• summary findings of research

• raw scores are not shown in table

What kind of Score is shown in Table

• must be converted to descriptive stats

• to present overview of results

• mean and standard deviation commonly used

• as measures of central tendency and dispersion

What are Bar Charts

• type of graphical display

• deals with categorical data not in particular order

• categories order doesn’t necessarily matter

• e.g. cbt as one bar, no therapy as another

How are Bar Charts presented

• data shown on x-axis is discrete

• data shown on y-axis is score/percentage

• do not have gaps between categories

What are Histograms

• continuous data version of bar chart

How are Histograms presented

• x-axis represents categories that have been measured

• e.g. number of marks on exam across year group

• y-axis represents frequencies of each category

• e.g. frequency of q5 being awarded full marks

• do not have gaps between bars

What are Scattergrams

• display results of correlations

• shows point at which two separate pieces of data meet

How are Scattergrams presented

• co-variables presented along x-axis or y-axis

• strong positive correlation will be shown regardless of which axis is chosen

• arrangement of points on scattergram will indicate correlation

What is Distribution

• spread of data around the mean

• for specific sample or population

• researchers are interested in extent to which one data set varies from the mean

Normal Distribution

• symmetrical around mean

• most scores close to it in centre

• peak in the middle where mean value is located

• bell curve

• extreme outliers fall within tail ends of curve

• tail ends never touch x-axis as no assumption is made on data

What can Normal Distribution be used to test for

• signs of deviance from the norm

• e.g. people who score beyond two standard deviations of mean may rank as having extreme high or low

How does Normal Distribution have Central Tendencies presented

• mean mode and median

• all appear at peak of curve

• scores to left = less than mean

• scores to right = more than mean

Skewed Distribution

• one tail is longer than other

• asymmetry in graph curve

• two halves do not mirror each other

• data is not distributed equally on both sides

• mean is most affected by skewed distribution

Positive Skew

• most values are found towards left side of graph

• long tail on the right

• mode (highest frequency)

• median (mid frequency)

• mean (lowest frequency)

Negative Skew

• most values found towards right side of graph

• long tail on the left

• mean (lowest frequency)

• median (mid frequency)

• mode (highest frequency)