Short course
Advanced Excel for survey data analysis
A survey is expensive to carry out and yields little if the analysis stops at a few percentages copied by hand. Yet that is what happens most often: the raw collection file goes unused, tables are redone one by one with every new version of the data, and results arrive too late to be useful for decision-making.
- Duration
- Short module, 5 sessions of 3 hours, 15 hours
- Format
- In person or online
- Credential
- Certificate of attendance
Tuition75,000 XOF
By the end of the programme, you will be able to
- Understand the structure of a raw file from digital or paper data collection
- Clean a dataset: duplicates, impossible values, inconsistencies, non-responses
- Standardise and recode free-text answers
- Handle multiple-choice questions and filter questions correctly
- Build a codebook, the reference document for any analysis
- Draw up a tabulation plan before producing a single table
- Produce accurate count and percentage tables, with the right denominator
- Build cross-tabulations disaggregated by sex, age, area or status
- Apply weighting and understand what it changes
- Choose and build the right chart for each type of result
- Write the interpretation of a table without over-interpreting differences
- Automate table updates when the dataset changes
- Use artificial intelligence to recode, code open-ended answers and draft a first analysis
Who it is for
Target audience
The module is for those who process survey data: monitoring and evaluation officers, enumerators and field supervisors, research officers, programme officers, NGO and association managers, statistics office staff, consultants, students and PhD candidates conducting field surveys.
Prerequisites
Two requirements:
- Be able to use a computer and have opened an Excel workbook before, even without mastering it
- Have a laptop, or rent one from the centre for the duration of the module
No prior statistical training is required. Each participant is invited to bring the file from a real survey, even raw and incomplete: the results chapter produced in class is based on their own data. A complete survey dataset is provided for those who do not have one.
Detailed program
5 sessions · 15 h of teaching
- What a digital data collection export really contains and how it is organised
- Export formats and clean import into Excel without distorting the data
- The expected structure of an analysis dataset: one row per respondent, one column per variable
- Spotting duplicate records and dealing with them
- Detecting impossible values, internal inconsistencies and empty questionnaires
- Non-responses, refusals and missing values, and coding them explicitly
- Standardising free-text answers: case, accents, spaces, multiple spellings
- The text functions essential for cleaning
- The cleaning log: recording every correction made to the raw data
- The golden rule: never modify the raw file, always work on a working copy
- Exercise: clean your own dataset and produce your cleaning log
- Recoding variables: age groups, grouping categories, binary variables
- Handling multiple-choice questions correctly, the most common mistake
- Filter questions and skips, and the denominator to use for each question
- Building derived variables, scores and composite indices
- The codebook: name, label, categories, codes, filter, source
- Readable labels and replacing numeric codes with labels
- The tabulation plan: which question, which cross-tabulation, which expected table
- Consistency between the survey objectives and the tables to produce
- Structured tables and dynamic ranges that update themselves
- Exercise: produce the codebook and tabulation plan for your own survey
- The frequency table: counts, percentages, valid percentages, cumulative percentages
- Choosing the denominator, the question that determines the accuracy of the whole table
- Pivot tables applied to a survey dataset
- The direction of percentages in a cross-tabulation: row, column, total
- Systematic disaggregation by sex, age, area and status
- Means, medians and standard deviations by subgroup
- Slicers and timelines to explore the dataset live
- Weighting: why it exists, how to apply it in Excel, what it changes
- Counts that are too small and the rule of caution before publishing a percentage
- Standardised formatting of a results table for a report
- Automatically generating the complete series of tables in the tabulation plan
- Exercise: produce the first twenty tables of your tabulation plan
- Choosing a chart by type of result: pie, bar, stacked bar, histogram
- Charts to avoid and misleading representations, intentional or not
- Readability rules: titles, units, counts displayed, no clutter
- The colour palette and visual consistency across a whole report
- Comparative charts between areas, between groups and between survey rounds
- Writing the interpretation of a table: the figure, its reading, its scope, its limit
- Not confusing a real difference with one due to sample size
- The structure of a results chapter and the sequence of tables
- The executive summary and key messages for decision-makers
- Exporting tables and charts to Word without losing formatting
- Dynamic links between the workbook and the report, for automatic updating
- Exercise: produce your complete results chapter, with tables, charts and text
- What artificial intelligence brings to survey processing, and its limits
- Confidentiality of survey data and anonymisation before any assisted processing
- Having a complex recoding formula written and explained, then checking it
- Having a column of free-text answers typed twenty different ways standardised
- Coding open-ended answers by theme: method, checking, correct classification rate
- Having the coding frame for an open question produced, then adjusting it by hand
- Having inconsistencies and suspicious questionnaires detected in a dataset
- Having a first draft interpretation of each table written, to be systematically reworked
- Producing the executive summary and key messages from your own tables
- Adapting the presentation to three audiences: management, donor, surveyed community
- The mandatory check: recalculating every figure yourself before publication
- Automating collection through an online form and updating the analysis workbook
- Final exercise: finalise your results chapter and present it to the group
- Presentation of attestations
Assessment and certification
The training attestation is issued to participants who have attended at least four of the five sessions and submitted their finalised results chapter. It states the module title, the 15-hour duration, the dates and the skills acquired, and carries a unique number with a verification code.
Participants also receive:
- The analysis workbook and results chapter they produced
- A model survey analysis workbook, from the raw file to the final tables
- A codebook and tabulation plan template
- A complete survey dataset to practise with after the course
- The complete course material in PDF, session by session
- A quick reference of Excel formulas useful for survey processing
- A library of ready-to-use prompts for artificial intelligence assistants
Upcoming sessions
Next dates being prepared
Leave your email: we will let you know as soon as registration opens.
Frequently asked questions
75,000 XOF per participant. Training for a team or an organisation is quoted on request.
5 sessions of 3 hours, 15 hours.
Yes: you can attend onsite or online, as you prefer.
Two requirements: be able to use a computer and have opened an Excel workbook before, even without mastering it ; have a laptop, or rent one from the centre for the duration of the module ; no prior statistical training is required. Each participant is invited to bring the file from a real survey, even raw and incomplete: the results chapter produced in class is based on their own data. A complete survey dataset is provided for those who do not have one.
A certificate of attendance is issued at the end of the course.
Fill in the online enrolment request: no payment is taken on the site. The firm reviews your application, then contacts you to confirm your place and the payment terms. No session is scheduled yet: leave your email on this page to be told when registration opens.
Join the next cohort
Cohorts are deliberately small to guarantee genuine individual support. Reserve your place or ask us your questions.