## 3. Data transformation and presentation

As described in Module A3, the first step to use EFA data is to transform them into EFA indicators, summary tables and charts that school managers, education administrators and stakeholders can easily understand, interpret and use as a basis for increasing understanding and making decisions.

In this section, we will learn about four common ways to present data. These are: **summary lists; tables; charts; and text.**

Once data and information are recorded and available at school, descriptive statistics can be produced to summarize the data, to describe the situation, and to identify issues and factors. We can produce descriptive statistics by:

**sorting and re-grouping data****transforming raw data into indicators such as percentages, rates and ratios.****presenting the data and indicators in tables, charts and texts which enable easy analysis, interpretation and use.**

Descriptive statistics also include summary statistics such as averages, range, median, mode, and standard deviation. These summary statistics can help people to understand the nature and characteristics of the data set and the phenomenon, which is important when they are analysing and interpreting data and indicators in order to understand the situation and to make decisions. More and more, charts and other graphical presentations are produced to present education data and indicators, using statistical and data presentation tools.

## 3.1 Basic Principles

When preparing a presentation of data and indicators, we should ask ourselves the following questions:

**What am I trying to communicate?****Who are my audience?****What kind of presentation will be most effective?****What might prevent my audience from understanding the data in the presentation?**

Decisions about the kind of presentation and the tools to use depend on the main objective of the presentation and the target audience. each presentation tool has its own characteristics, advantages and disadvantages, but the choice will depend upon the context within which the presentation is being delivered, and the message to be conveyed. for example, a chart may highlight key patterns but fail to show the details. it should be presented together with a table that contains the detailed data. descriptive text accompanying the chart and table can draw attention to salient findings and possible causes and implications. **Consistency of style and logical flow of ideas are vital to a good presentation.**

The goal of a presentation is to make the reader want to see it, analyse it, and understand the key messages with minimum effort. A good presentation is therefore:

**informative****self-explanatory****pleasant in appearance****easy to understand, interpret and use**

Always try to communicate with your audience in the most simple, direct and efficient manner possible. do not overload presentations with too many tables, charts and text. identify the key findings and ideas, and then present them in the simplest possible way.

## 3.2 How to use summary lists to present data

In school, summary lists of students by class can be used to manage the classes throughout the school year.6 Sometimes, however, the default way of presenting student lists by alphabetical order or by seating order may be less efficient to use for specific purposes.

for example, it may be helpful to** sort or rank the students in a summary list** using criteria such as by the number of days absent or their scores in subject tests. The sort order can range from high to low (descending order), or from low to high (ascending order). **By sorting and ranking, we can easily identify at the two extremes of the sorted list, those students who are most and least frequently absent from class, or the students who achieved the highest or the lowest scores in a specific subject.** The values at the two extremes will also give us the maximum range of difference between them. Sorting and ranking are therefore useful data transformations that can facilitate analysis and interpretation of summary lists.

TIPS FOR SORTING AND RANKING A SUMMARY LIST:

- Clarify what you want to know (e.g. the highest and/or the lowest; grouping people/objects with common characteristics; ranking within sub-groups, etc.)
- identify the data series to be sorted/ranked
- Decide on the sorting/ranking criteria and order (i.e. ascending or descending)
- Repeat steps 2 and 3 if you want to further sort/rank within sub-groups
- Perform the sort and verify the results. Change the sorting data series and/or criteria to analyse other aspects.

Other summary lists at school may include lists of classes, teachers, equipment and textbooks which can be sorted to compare performance and to identify management issues. district and provincial education offices, and the Ministry of education can maintain updated lists of schools which list the name of school, number of grades, classes, students, and teachers, together with indicators such as pupil-teacher ratio, intake rate, enrolment ratio, repetition rate, dropout rate, percentage of under- qualified or untrained teachers, and other essential management information and reference data. These data can be used to sort and rank the schools in order to identify those which perform well and others which under-perform or have problems with specific aspects.

### Activity 2

#### Review and discuss with other colleagues about practices in preparing, sorting and ranking summary lists, and then answer the following questions:

- What kind of summary lists do they produce? Are these lists usually sorted and ranked? If yes, using what kind of data as sorting criteria? if no, why not?
- What kinds of summary lists should be produced in the future in your school/district/ province/ ministry? How should they be sorted and ranked?

## 3.3 using tables to present data

As discussed in Section 7 of Module A1, summary tables can present:

**counted numbers****totals and sub-totals****percentages****ratios****averages****rate of change**

We can also use tables to present other types of derived statistics and indicators so we can com- pare the relative sizes, range, means, and spread. Presenting data in tables can help to analyse the underlying patterns and differences, which can lead to better understanding and interpretation of the data.

Summary tables are different from summary lists. They are mainly used to present counts of objects, items or phenomena, such as the number of teachers by qualification or the number of students by sex, age and grade. Summary tables can be simple one-dimensional tables with one line (or one column) of data, like in example 1a below that shows the number of teachers by qualification in a primary school. in this example country, all primary school teachers are expected to have at least a post-secondary diploma.

It can be seen that seventeen teachers in this school are qualified; four are under-qualified; and two have qualifications above the requirement. Sometimes it may not be easy to interpret and under- stand these absolute numbers, especially if we are comparing several schools of different sizes. if this is the case, we can add a second column (shaded) to the table to show the percentage distributions of the values, as in example 1b below.

We can see that this new column of percentage distributions enhances our ability to interpret this table. for example, the following key messages can be highlighted: (a) 74% + 9% = 83% of the teachers possess post-secondary diploma or university degree; (b) 0% + 4% + 13% = 17% of the teachers are under-qualified; (c) actions have to be taken to upgrade the qualifications of 1 + 3 = 4 teachers.

When needed, the tables above can be further expanded to become two-dimensional (or even multi-dimensional tables) to allow for even more detailed and in-depth analysis of teacher qualifications. in example 1c below, we can see additional columns of data and percentages have been added to the same table, separately for male and female teachers to show how qualifications vary by gender (shaded).

Multi-dimensional tables can enable more dynamic and detailed analysis of data and indicators, both horizontally (by rows) as well as vertically (by columns). This provides a more detailed pic- ture and understanding of the situation we are studying.

In the ‘% of female teachers’ column in the middle of example 1c, we can see that almost two-thirds (65%) of the teachers in this school are female, and the proportion of female teachers with post- secondary diplomas is even higher: 71 per cent. This is further confirmed in the last columns, which show that overall, female teachers are more qualified than their male counterparts, with 80 per cent having post-secondary diplomas as compared to 61 per cent for male teachers.

Quite a few computer software packages offer flexible and easy-to-use functions for constructing, handling and analysing numerical tables, especially multi-dimensional tables such as the PivotTable function in Microsoft excel. Some national statistical websitesoffer even more dynamic data retrieval and tabulation functions online.

Based on past practices and current needs for information, school managers and education officers at higher levels can each create a set of standard summary lists and tables that are frequently used, and update them on a periodic basis to support planning and management functions.

The design of some of these summary lists and tables may correspond to the tables and lists in the annual school census questionnaire. in this way, the data can be directly copied into the corresponding school census tables. The same applies to producing standard tables and lists for the annual school report, if such reports are required.

**TIPS FOR PREPARING AND INTERPRETING MULTI-DIMENSIONAL TABLES:**

- Large, complex multi-dimensional tables can be difficult to interpret and use. Avoid cramming too much data into a single table. Split it into several simpler tables.
- Avoid using more than two layers of headings in the horizontal and vertical headings.
- Group rows or columns together where data are related and can be directly compared and interpreted, to simplify analysis and interpretation.
- Place derived percentages, rates or ratios next to (or under) the original data so that they can be analysed together. if there are too many derived indicators, re-group them into a separate part of the table using the same group headings, or into separate tables.
- Start analysis and interpretation along each single dimension, and then combine and synthesize the findings to develop a more thorough understanding of the situation.

### 3.3.1 Types of tables

There are two main types of tables: **reference tables and investigative tables**. in this section, we examine both types of tables.

**Reference tables**

Reference tables allow users to make the best possible use of detailed raw data. They often appear in the annex of technical reports, and can also be made available in electronic format. example 2 is a reference table that presents the number of students enrolled in a school by gender, age and grade.

in example 2, we can see that the reference table shows in most cells, absolute numbers of students detailed by single years of age, grade-by-grade and separately by sex (female [f] and male [M]).

This is an important reference table in primary education because it enables in-depth analysis of the patterns of participation by presenting data about the distribution of students by age and by grade, and the age-grade correspondence of students by gender. The absolute numbers presented in the table can be used to make additional calculations to derive indicators such as the age-specific and grade-specific enrolment ratios, and the percentage of over-aged and under-aged students in each grade which are crucial parameters in managing the universalisation of primary education.

By summarizing the calculated indicators and the results of further analysis of detailed data in the reference tables, we can create investigative tables which present the key findings for use by non- technical audiences.

**Investigative Tables**

Investigative tables like example 3 below present only the key data and indicators that can be under- stood and interpreted by lay persons. They are often accompanied by descriptions and explanations which highlight the findings, explain the terms used, and ways to interpret the data. numbers, for example, are often rounded to whole figures and supporting information and indicators such as percentages and averages are provided to help people make sense of the data. Besides highlighting and presenting key facts, figures and issues, investigative tables also lend themselves to further analysis.

**Type of table to use**

Tables offer a structured way to organize and present detailed data for further analysis, synthesis and interpretation. The main disadvantage of using tables is that the readers only see numbers, which require specific aptitude and skills to correctly understand and use them.

Clarifying beforehand the objective of the presentation and the profile and needs of the target audience can help us to select the right type of table to use. To illustrate, investigative tables are probably the best choice to accompany a newspaper article for the general public, whereas a reference table would be more appropriate for a technical presentation given to a group of analysts and statisticians.

### 3.3.2 Basic components of a table

A properly presented table should include a title, headings, rows and columns of data, units of measurement, degree of accuracy, and footnotes indicating the source of data and specific data characteristics.

**Title**

The title is the main description of the table. it should be concise and, for the sake of interpretation and record keeping, be informative and meaningful. The title should include a date or year of reference regarding the data, the place such as a country, regions, provinces, districts, villages or schools, and any other attribute that is common to all the data entities in the table (see ‘Title’ in example 4 below).

**Headings for rows and columns**

Each row and column needs a heading that describes the data in that row or column, using labels such as ’number of female students’ or ‘% of female students’. Headings usually vary from row to row and from column to column so as to distinguish between them. They may be re-grouped under a major common heading (see example 4).

**Unit of measurement**

A unit of measurement should be specified for every entry in the table. The unit of measurement is usually given in the title (if the same unit of measurement applies to the whole table), or in the column or row heading (if rows and headings use different units of measurement). Precise use of measurement units is essential for the correct interpretation of data. different units of measurement can have different interpretations. To facilitate understanding and interpretation, it is advised to follow the convention used within the country. each EFA indicator also has a specific unit of measurement such as percentages, ratios or rates (see example 4).

**Degree of accuracy**

Numbers may be rounded to a nearby value using a degree of accuracy that is appropriate to achieve the goal of the presentation. When rounding numbers, unnecessary trailing zeros should be removed and the units of measure altered accordingly. Rounding can be done either:

- To a certain number of digits, such as to the nearest thousand (so 45,647 becomes 46,000) or to the second decimal place (so 45.647 becomes 45.65).
- To a certain number of significant, non-zero, digits. for example, to the 2nd significant digit (so 343,833 becomes 340,000 and 4,564 becomes 4,600).
- In some cases, we may want to write 46,000 as 46 and note the recording convention (‘numbers in millions’, or ‘numbers in thousands’ or ‘000’) in the title or appropriate row or column heading (see the unit of measure in the first column of example 4). The same principle applies to decimal points for percentages, rates and ratios.

**Footnote**

Footnotes provide important information that helps us understand the data in the table. The following information is often provided in footnotes:

**source of the data which enables interested readers to pursue extra information and estimate the quality of the data****conventions used and further explanations of terms found in the table****complete notation of row or column headings or the title if the labels are too long to print in full in the main body of the table****differences in status of some entries in the table****data limitations; and/or****any other exception/deviation from the stated norm (see ‘Footnote’ in Example 4).**

**Layout**

A clear, well-structured layout makes it easier for the reader to interpret and understand the information in the table. One popular style is the two-way layout like the earlier example 1c. normally, the table includes factor A and factor B. in example 5 below, factor A is a list of regions. factor B is about the number of out-of-school children in 1999 and 2006.

### TIPS

**Font style:**different font styles may be used to highlight specific items of a table that require special attention.**The ordering of rows and columns:**This is critical for clarity. As a general rule, rows and columns should be arranged following a natural or logical order, ranked, for example, by alphabetical order, geographical location, year or magnitude. Alphabetical and geographical ordering are both useful for reference, whereas ordering by magnitude makes the ranking of the different entities immediately obvious.**Numbers:**numbers are easier to compare when the table has a vertical orientation. The human eye can make comparisons more easily when reading down a column of data than across several columns.**Consistent appearance:**The appearance of the table should be kept consistent through out the report. for example, conventions for labelling and ordering rows and columns should be kept consistent as much as possible. A common mistake is to switch or mix up the rows and columns and label- ling across tables within a report.**Number the table:**Tables should be properly numbered for ease of reference.**Unnecessary distraction:**Avoid designing complex tables with many layers of headings for row and columns. Break them into smaller tables. Many tables that should logically appear together should be placed in the appendices.

### Activity 3

#### Review and discuss with education officers, school managers and personnel about their practices in creating and using summary tables, and then answer the following questions:

- What kinds of tables are produced in your school/district/province/ministry? for what purpose are they produced?
- What difficulties have you encountered while designing, producing and interpreting tables?
- What other summary tables can be produced and used in your school/district/ province/ministry?

## 3.4 using charts to present data

Charts can be easier to understand than tables, and are often more effective for highlighting important information within large data sets for readers who are less apt with numbers. **Charts are especially useful for showing patterns and comparing trends**, and since charts are more visual and attractive than tables, they can help in presentations. On the other hand, **charts are not suitable for communicating more detailed and precise data, and can be time-consuming and expensive to design.**

Charts and graphs have the following advantages. They:

**are more attractive and easy to understand than tables****enable the reader to ‘see’ patterns in the data****are easy to use for comparisons and analysis****allow information to be presented in various ways using different types of charts allow for special designs like demographic pyramids and thematic maps**

The main purpose of charts or graphs is to visually impart information that cannot be easily read and interpreted from a table of data. it can be sometimes difficult to ’see’ patterns, trends and contrasts in a table that has many data points. using charts can be a good method of showing trends and changes in statistical data. Charts can also be used to make predictions and forecasts, and to compare two or more data sets.

### 3.4.1 Basic components of a chart

Similar to tables, charts should have a title, axis labels (including the units of measurement), tick marks on the axes (with labels for some tick marks and subgroups), and footnotes and references to the source data.

When designing a chart, one should take into consideration the order in which the reader looks at the elements of a graph. By virtue of reading habit, our eyes tend to first look at the title, then move to the graph itself, and finally to the explanatory labels and footnotes, before coming back to study the overall pattern and details in order to understand the key messages.

The explanatory title, labels and footnotes are therefore essential for understanding and interpreting the chart. One must nevertheless bear in mind that too many text details and labels can distract the reader from the main message.

An effective chart also has the following characteristics:

**Clear objective and key messages to be presented****Good choice of graph type for the information that is to be presented****Appropriate level of simplicity or complexity, depending on the readers’ abilities to analyse, interpret and understand**

### 3.4.2 Types of chart

There are many different types of charts, including:

**Pie chart****Line chart****Bar chart****Area chart****Scatter plot**

each of these is described below.

### Pie chart

**Pie charts are useful for illustrating percentage distributions of components within a total** and are therefore suitable for presenting overviews of data. The breakdown of the annual edu- cation budget by types of expenditure is a good example of information that can be presented clearly using a pie chart (see example 7). The area of each slice of the pie is proportional to the relative share, or frequency, of each item that makes up the total.

**The pie chart however is not suitable for displaying data with a high degree of accuracy**. This is especially true when an item represents a very small share of the total, in which case its slice may be very thin and difficult to measure. it can also be difficult to see the difference between two items with similar but slightly different shares of the total. it is usually necessary to clearly label the value of each slice of the pie in order to overcome this limitation.

**Do not include too many data items on the same pie chart.** Too many slices, especially thin slices or slices of a similar size, may make a pie chart difficult to interpret. A maximum of 5 or 6 slices is optimal.

**One pie chart can only show the distribution of data items at one point in time.** If, for ex- ample, we want to show how the distribution of the education budget changed over two or more years, we need to create a separate pie chart for each year and place them next to each other to allow comparisons to be made. The 100% stacked bar chart (described later in this section) is a more appropriate chart for displaying this type of proportional data from a time-series.

### Line graph

**A line graph is a simple way to show time-series or trends.** A single graph can display multiple lines, with one line for each data series or sub-population, such as the number of students by sex, or gross and net enrolment ratios for boys and girls over time (see example 8). **Horizontal grid- lines** make it easier for the eye to follow the plotted lines, to identify the data values along the data series, and to get a clearer picture of the trend and changes in the values for different data series. It is **generally not advisable to label every data point** in a line chart, unless it is important to show precise values that are quoted in the accompanying text.

### Bar chart

A bar chart displays data as rectangular bars with lengths proportional to the values that they represent. **Bar charts are useful for comparing two or more values,** such as the gross enrolment ratios by sex and region in Viet Nam (see example 9). The bars ca be vertically or horizontally oriented (see example 10). Horizontal bar charts are often used when there are many bars and each bar requires a long label, which would overlap or be truncated in vertical bar charts.

**Stacked bar charts** are used to present values that have components. example 11 presents the official primary school age population by region in Viet nam in 1990-1991, by stacking the population size of boys for each region on top of that of girls. The entire length of each bar indicates the total primary school age population in each region. The lengths of separate component bars for boys and girls (within each bar) show both their absolute sizes as well as comparison with each other.

**The 100% stacked bar chart is especially useful for comparing percentage distributions,** each of which adds up to 100%. Such percentage distributions can refer to different regions or schools within a district, or show changes over time. example 12 shows the gradual shift in the shares of education expenditure from 1990 to 2000.

### Area graph

Area graphs are an alternative to the stacked bar chart (see example 13). We create an area graph by drawing lines to connect the values of component data series, and then stack them one on top of another. **Area graphs are particularly effective for showing changes in total (and component) values over time.** in example 13 we can see consistent increases in the estimated number of male and female literates in the world. At the same time, the graph shows that the estimated number of illiterates decreased only slightly over the period from 1980 to 2015. When designing an area graph, we must make sure that the time scale is drawn correctly, using a linear scale for the time points.

### Scatter-plot

**Scatter plots are used for showing the relationship between two variables, or data series, by plotting pairs of values.** in example 14, the apparent intake rate for boys is plotted against the intake rate for girls for each region of Viet nam in 1999. On this graph, we observe that almost all of the plotted points fall below the diagonal line. This shows that boys generally have less access to pri- mary education than girls in Viet nam.

With recent improvements in computerized graphic technologies, it has become very easy to pro- duce **3-dimensional (3D) graphs** (see example 15). Compared with flat 2d graphs, 3d graphs can show additional data series in the same graph, thereby allowing for more sophisticated comparisons and analysis. 3d graphs with many data series can however become overly complex and difficult to interpret, especially when they are not well conceived and designed.

### Activity 4

#### Review and discuss this section about charts with education officers, school managers and personnel. Find out about their practices in creating and using graphs, and then answer the following questions:

- Does your office/school frequently produce graphs? if yes, what kinds of graphs are produced? How useful are these graphs?
- Do you use a computer to produce graphs? Which computer software and functions were used in producing tables and graphs?

## 3.5 Selection of data presentation

How do we decide whether to present data as a table or as a chart?

In this section, we look at the main differences between tables and charts and suggest some criteria for choosing between them:

**Tables, with their columns and rows of information, interact primarily with our verbal system. We process information in a sequential fashion, reading down columns or across rows of numbers, comparing this number to that number, one pair at a time.****Charts are perceived by our visual system. They give shape and form to numbers. To see patterns and relationships is a natural function of visual perception. A single chart can convey important features of the data more vividly and memorably than columns of data.**

### TIPS

- Tables work best when the data presentation:
- contains many data series for cross-analysis
- is used to look up or compare individual values
- requires precise values
- contains values that involve multiple units of measurement

### TIPS

- Graphs work best when the data presentation:
- is used to identify patterns and trends
- is used to communicate a message that can be highlighted in the shape of the graph
- is used to visually reveal relation- ships among several values

As explained in Sections 3.3 and 3.3.1 above, always start by organizing and presenting data in table form to study the basic structure and inter-relationships among the data series. next, calculate percentages, ratios and other indicators that can help to identify patterns, differences or changes, and show these next to the absolute numbers. Graphs may then be used to visually illustrate the data, or to draw attention to significant patterns and changes in order to facilitate understanding of their significance.

By selecting the most appropriate type of chart to present one or several data series, graphs can be used to visually analyse and identify patterns and significant highlights which are not obvious when looking at a table. if needed, graphs may be shown together with tables so as to allow the reader to see the detailed data and indicators.

#### Activity 5

#### Review and discuss with education officers, school managers and personnel about their practices in creating and using tables and graphs, and answer the following questions:

- Do you create and use tables and graphs to present information about your school/district/ province/ministry? When and for what purposes?
- What difficulties have you encountered in selecting what tables and graphs to produce and use?
- How do you determine when to use tables or graphs to present data and information?
- What have you learned about using tables and charts to present data?

## 3.6 Time-series data

Sometimes, we need to present information about changes over time. To present time-series data, we use summary tables and charts that effectively show changes in values from one time period to the next in chronological order. We can, for example, show how the number of students and teachers change over a number of years (see example 17). As long as the definition of the data and the collection methods were consistent over time, we can use tables and charts to study past trends and estimate future values.

Example 17 shows two time-series lines in the same graph – one for students and one for teachers. The two time-series have been combined, with different scales on the left and the right side of the graph. it is interesting to see that the two trends generally overlap but the lines crisscross over each other. This can be interpreted as overall correspondence in the availability of teachers in relation to the number of students over the years, although there have been noticeable fluctuations, or instability, in the teaching force. Additional analysis using the calculated indicator of pupil-teacher ratios (PTRs) in the right-hand column of the table can help to further explain the situation and trends.

if we look closely at the student and teacher trends in the table above , the PTRs calculated for each year indicated that such ratio varied from a low of 33 to a high of 42 over the years. To begin our analysis of this situation, we should first check the norms for PTR specified by the Ministry of education. if the national norm is 35 pupils to one teacher in primary school, we can observe that the school under review was only within the national norms in the years 2001, 2003, 2006 and 2008. for the other years, the PTR is higher than 35, which indicates that fewer teachers than the national norm were employed to serve the student population in the school. This calls for measures to be taken to ensure that enough teachers are hired or assigned to the school to keep within the national norm for PTRs.

Within a school, PTRs may be calculated for each class so as to identify disparities between classes and also identify the classes in which teachers have an exceptionally heavy workload. Appropriate remedial measures can be adopted to solve such imbalances in teacher workload, or to compen- sate teachers who have a heavier workload. in doing so, we must take care to ensure that classes with students who have special needs and subjects that require more intensive student-teacher interaction maintain a PTR which is appropriate to the needs of the class, but which may be below the national norm.

### Activity 6

#### Review and discuss practices for analysing and interpreting time-series data with education officers, school managers and personnel, and then answer the following questions:

- Why are time-series data important?
- How regularly are time-series produced, analysed and interpreted in your school, district, province or ministry? What type of time-series data are produced? How?

## 3.7 using text to present data

**Tables and charts are often accompanied by descriptive text that highlights the findings, pat- terns, issues and implications of the data.** Textual descriptions and discussions play a crucial role in almost every kind of data presentation, especially for people who are not familiar with data tables and charts. Many people even prefer plain textual descriptions to tables and charts, or at least need some clear and simple explanations to help them understand the data and important points presented in the tables and charts.

**Basic rules**

One of the important functions of text is to provide a verbal description of the data in tables and charts. We should remember **five basic rules** when drafting a text to describe a table or chart:

**Try to capture the readers’ interest**– While staying within the confines of scientific rigour, the writer should strive to enliven the text by highlighting key findings and meanings.**Take time to write clearly and succinctly**– draft and re-draft to clearly and unambiguously describe the data.**Ensure consistency of language and style throughout the report or presentation**– Often sections of a single piece are written separately for each table or chart, so a final check to ensure consistency is advisable.**Avoid unnecessary repetition**– if parts of the report or presentation are written separately, contents get repeated. Review the written text to eliminate unnecessary repetitions and to harmonize the texts.**Focus on the main points and minimize unnecessary details**– Present the most important information first, and add details only if absolutely necessary.

A verbal summary should simply accompany the table or chart to explain what the data reveal. it should not dwell on issues that are too specific or too detailed. nor should it repeat what is obvious in a table or chart unless there is a need to emphasize the importance of a certain aspect or limitations of the data. Sometimes, a verbal summary is all that is included in a presentation, particularly when the findings are so simple that any other summarized display like tables and charts are not justified. Or, when numerical or graphical presentations are too complex, it is better to include them separately in the appendices. The following are some additional basic rules:

**Keep the summary short**– never allow the verbal summary to expand into an itemized account of each entry in the table or chart. Position the summary in the text close to the table or chart to which it refers. Quoting directly the key reference numbers is the best way. for example: “From Table 2 we can see that 46 per cent of girls (6-15) were out of school in 2001.”**Use ‘emotional’ descriptions and wording sparingly**– Sensational messages can be effective with a non-technical audience, but they can communicate biases or lead to biased interpretations. for example: “Education expenditure per student in China rose by 10 per cent” may be better than: “Education expenditure per student in China shot up by 10 per cent!”.- Unless writing specifically for expert readers,
**avoid using unnecessary technical terms.** **Be cautious when attributing causality to a factor**– for example, some erratic movements in a data series may be due to changes in definitions or measurement unit, rather than actual changes in the underlying event. do not jump to conclusions that ‘this caused that’, unless there is ample evidences to justify it.

### TIPS

- Avoid long sentences. Short, sharp sentences are more effective.
- Paragraph breaks provide the reader with a short rest, and are necessary to maintain the reader’s interest. On the other hand, having too many short paragraphs may ruin the flow of the text and make it confusing and uninteresting.
- Commas provide pauses within a sentence, but, if over-used, may be bothersome and misleading. Minimize the use of brackets. When several pairs of brackets are used in the same sentence or sec- tion, the reader may become confused and lose track of the central theme.
- Avoid repeating the same wording in close proximity. it looks careless and may be distracting. To avoid this problem, try to search for and use synonyms.
- Avoid unnecessary words and phrases. These give the reader extra work and may distract them from the central theme or main findings.
- Keep the language simple. Readers are usually more impressed by clear language than by words that they do not fully understand.
- Be logical. ensure that the structure is sequential and that the conclusions reflect the body of the text. This reduces the chances of the readers becoming confused.
- Ensure that the use of articles is clear. Whenever words such as ‘it’ or ‘that’ are used, be sure that the reader has no doubt about what these pronouns refer to.
- Adopt conventions and keep to them throughout. Conventions can include headings, numbering, use of abbreviations and acronyms, or other matters of style such as the use of digits and written numbers. if necessary, these should be spelled out by the writer or writing team. if conventions are maintained throughout, a report will be clearer and better received.

## 3.8 Other tools to present data

### Maps

**A map is an abstract, simplified representation of a geographical area that is drawn to scale to highlight relationships between objects and indicators within the drawn space.** Maps are usually two-dimensional representations that use short texts, signs, symbols, colour-codes and many other graphical conventions. **Maps are an excellent medium for displaying the distribution of data and indicators across geographical areas.**Within the context of EFA monitoring, maps can be used to show the location of schools in different regions and localities. each school may be labelled with education indicators, such as the school’s enrolment ratio, PTR and education expenditure per student. Maps can also be used to show the location and areas of concentration of adult illiterates and school-age children who are not in school. Softwares like Devinfo and other GiS (Geographical information System) have automatic data handling, plotting functions and extensive map portfolios.

### Photos, video and other media

Some data and indicators may be presented visually using photos and/or videos of schools, classes, students, teachers and educational activities. Other media, such as audio commentaries about progress, issues and future plans may also be used to enhance a presentation. We should, however, use multimedia elements judiciously and only when it fits with the occasion, or with the right target audience.

### Webpage

The internet is often used these days to present and disseminate data and indicators about education. Some national ministries of education have included education statistics and indicators in the form of summary tables, charts and texts on the Ministry’s website. The quantity and variety of such presentations are increasing with time, and becoming more and more attractive and easy to under- stand through the use of vivid colours, 3d graphics and other new technologies.

Some websites allow users to generate customized tables of data and indicators for various geo- graphical regions and time periods. users enter specifications about the type of data they require into a form to query the database, which in turn will generate the table and chart with the results.

Some schools have set up their own website to publish information about the school. They use tables, charts and texts to present data and indicators about the school.