Unit 3.2 Flashcards
Data is simply narrated, story fashion
Narrative or Textual
A systematic arrangement of data presented
in columns and rows for the purpose of comparison
Tabular
Pictorial representation of data
Figure or graphical
Data is presented in the form of a sentence/ Paragraph
Textual presentation
Most basic way of presenting data
Textual presentation
Characteristics of a good figure/ table
1 simplicity
2. Clarity
3. Continuity
• Augment rather than duplicate the text
• Convey only essential facts
• Omit distracting detail
Simplicity
- Easy to read – its elements (type, lines, labels, symbols,etc) are large enough to be read with ease in printed form
• Easy to understand – its purpose is readily apparent
Clarity
• Consistent with and is prepared in the same style as similar figures [and tables] in the same article
• Carefully planned and prepared.
Continuity
allow the amounts of raw data to be sorted and reorganized in a neat format
Tabular
results placed in an organized display of rows and columns that enables grouping of data by different classifications for comparison and better understanding
Tabular
allows the inclusion of only the most important or relevant data
Tabular
never put data in the table if you can describe it efficiently in one or two sentences
Tabular
Data in tabular form uses
- Summarizing and presenting data
- Data checking and editing
- Basis, aid in graph or chart construction
Strength of tabular form
- Easy to understand
- More compact and concise than textual form
- Present data in greater detail than a graph
- Can readily point out trends, comparisons or
interrelations
Parts of table
- Table number
- Table title
- Spanner
- Decked heads
- Table body
- Use arabic numerals in the order in which the tables are first mentioned in text
- Do not use suffix letters
- Partial or fractional numbering is unacceptable (i.e. no 5a or 5.1)
Table number
- Brief but clear and explanatory
- Titles should adequately explain the content of a table without referring the reader to the text
- Single-spaced
Table title
- Column spanner
• singular
• covers several columns each with its own column head - Table spanner
• located in the body of the table in order to divide the data in a table without
changing the columns
• covers the entire width of the body of the table
• used to combine two tables into one provided they have
similar column heads.
• may be plural
Spanners
- stacked headings (column heads and spanner)
- used to avoid repetition of words in column headings
Decked heads
- Cell is the intersection between a row and column
- Decimal values
• Use a zero before the decimal point
• Do not use zero if the number can not be greater than 1
• Use same unit of measurement and number of decimal places within
a column - Do not include columns of data that can be calculated
easily from other columns - Double-spaced (rows)
Table body
• Notes
- General note
• Qualifies, explains or provides info relating to the table as a whole and
gives explanation of abbreviations, symbols as well as the source
- Specific note
• Refers to a particular column or row or individual entry
• indicated by superscript lowercase letters
- Probability note
• indicates the results of tests of significance
- Order
• General, specific, probability
- Single-spaced
Table body
Consideration in table construction
- Relation of tables and text
- Relation between tables
- Ruling
- Word/textual table
- Numerical table
- supplements the text
- every table should be cited/ referred to in the text
- discuss only highlights
- place tables close to where they are first mentioned in your text do not split a table across pages
Relation of tables and text
- standardize a particular style (same format, terminology, etc.)
- combine tables that repeat data (identical columns should not appear in 2 or more tables)
Relation between tables
- 3 horizontal lines (APA feature)
- no vertical lines (allowed in exceptional cases)
Ruling
Types of table
- Word/textual table
- Numerical table
- Relative Frequency Distribution Table
• One-way
• Two –way
• Multi- way - Statistical Table
- Master Table
- Dummy Table
Numerical table
• Tables that have textual data
• Serve the same purpose as any table - comparison
Word/textual table
- a relative frequency is a ratio of the number of observations in a statistical category to the total number of observations (proportion or %)
- a table which presents proportion data
- categorical variables
Relative frequency table
Presents data with respect categories of one variable only
One-way
data with respect to cross-classification of two categorical variables
Two-way
data with respect to more than two categorical variables
Multi-way
• tables that can present data of a
- descriptive statistics
- inferential statistics
eg. ANOVA, Correlation
- or both
Statistical table
• to facilitate tabulation in the absence of a computer
• shows the distribution of observations across several variables of interest in a given study
• each observation is cross-classified across the variables
Master table
- contains raw data
- where you can plot all data
Master table
• skeleton tables
• preview of tables to be presented
Dummy tables
- help researcher clarify instrument
- help protocol reviewer
- guides data analysts/ programmers
Dummy table
Usual errors in table construction
- Comparison based on absolute numbers
- Use of wrong denominators in computing for percentages
• visual representation of relationship between,
but not restricted to, two variables
Graphical presentation
a graph consists of two axes called the x-axis
(horizontal) and y-axis (vertical) and each
corresponds to one variable
Graphical presentation
- Simpler to read
- More attractive and appealing
- Strengthen emphasis on certain aspects of the data
- With a wide point of view of the data
- Effective tool in delivering a specific message
- Shows trends and patterns in a large data set
- Comparison could also be made more striking
Advantages of graphical presentation
Disadvantages of graphical presentation
• cannot show as many sets of facts (detailed
information) as may be shown in a table
• can only show approximate values
• require more time to construct
• may be used to misinterpret results
Pointers in Graph Construction
- Title or caption must be clear and concise.
- Use simple graphs (usually the most effective).
- Graph should be self-explanatory.
- Use legends appropriately.
- Be careful when plotting your scales to
avoid misleading the readers and properly label them. - Use a scale break to avoid placing the graph high up on the grid.
Not appropriate to use graph when
• Data are very dispersed
• Too few data (categories)
• Numerous data
• Data show little or no variation
• Shows breakdown of a group or total where the no. of categories is not too many
• Use the % for each category, thus the total always equal to 100%
• Apply different shades to each pie slice to differentiate the different groups
Pie graph
• Readily compares the percentages in each category by comparing the heights of the bars
• Use percentages or rates when the total no. of observations for the groups are not uniform
• When % are used, the sum of the heights of all bars must equal to 100%
• For qualitative variables, categories may be arranged by
- Magnitude of corresponding figures
- Natural ordering
- Example: Educational attainment, Birth order
Bar graph
usually used for discrete quantitative variables
Vertical bar graph
Usually used for qualitative variables
Horizontal bar graph
- A bar divided into smaller rectangles representing the parts
• Similar to a pie chart
• Area of each smaller rectangle is proportional to the relative contribution of the component to the whole
• Preferable over the pie chart in situations where the compositions of two or more groups are to be compared
• Different shades or colors can be applied to the components to emphasize differences between parts of the whole
Component bar graph
- Graphical representation of the frequency distribution of a continuous variable
• Important characteristic - no space between bars
• Vertical scale may show the absolute or relative frequencies
Histogram
• Plot of dots ( representing values of the variable) joined with lines over some period of time in sequential series
• Time series is shown along the horizontal axis while the variable values are shown along the vertical axis
• Portray trends, i.e., changes in the variable with time
Line graph
• Similar to the histogram except that
- Frequencies are plotted against the
corresponding midpoints of the classes
- Can depict more than 1 one distribution
• A closed figure
Frequency polygon
Best used to show clustering results
Scatterplot
• Show relationship between two quantitative variables
• Gives rough estimate of the degree of correlation between the variables
• Needed data to construct actual values of each variables of all the subjects
Scatterplot
Aka boxplot
box and whisker plot or diagram
a graph summarising a set of data.
box and whisker plot or diagram
It is a useful way to compare different sets of data as you can draw more than one boxplot per graph. These can be displayed alongside a
number line, horizontally or vertically.
- The shape of the boxplot shows how the data is distributed
box and whisker plot or diagram
Inappropriate construction of graph
- Thick grid lines - line graph
- Big spaces - scatterplot
- Little to no variation
- Dispersed data
- Very few data
Clustering of data in scatterplot is
Negative result
More scattered plot in scatterplot is
Good results
