Event Sequence Dataset

Introduction

For the research that I did for my PhD-project I had to collect large amounts of longitudinal data. In their book on Organizational Change and Innovation Processes Poole, Van de Ven, Dooley and Holmes (2000) describe the structure of the datasets in which they recorded data that was collected during the Minnesota Innovation Studies. Based on their instructions I created a template for Event Sequence Datasets (ESDs) using Microsoft Access.

You can dowload the ESD here: ESD Template File

If you would like to cite this software, you can cite the following paper, in which the ESD is described:

Spekkink, W. (2013). Institutional Capacity Building for Industrial Symbiosis in the Canal Zone of Zeeland, The Netherlands: A Process Analysis. Journal of Cleaner Production 52, pp. 342-355.

How it works

The ESD structure assumes that you record your data in the form of relatively brief qualitative descriptions of occassions – what Poole et al. (2000) describe as incidents. For example, incidents can be brief descriptions of actions or interactions. The ESD requires you to assign a time stamp to the incidents. The ESD will put the incidents in chronological order – see Spekkink (2013) for more details. The ESD structure also provides an interface for coding activities (both descriptive and categorical coding is possible). The ESD also allows you to keep a code book (including explanations of categorical codes) within the dataset itself, such that you have everything in one file (so make backups ;-)). The data can be exported as an excel-file. You can choose to export all the incidents in the dataset, or just those incidents that have been assigned a code.

You’ll need to have Microsoft Access installed to your computer to be able to use the ESD. If you ever want to change anything in the structure of the dataset, you can access the actual tables, forms, queries, and the code that I have written by opening your file, while holding the shift-button on your keyboard.

If you start using the ESD, it is best to make a copy of the template first. There are no save or load options. Once you start entering data into the ESD, that data will be stored. If you delete data, it is deleted permanently (again, make backups). This tool made my PhD-life a lot easier. I hope it can do the same for others.

Overview of interface

The first screen that you will see is the main menu from which all other relevant windows can be accessed and from where you can export data to excel. If you want to, you can enter a name for your case in the field below the top bar.
MainWindow

Although the Incident Entry Form is the most important function (the one that you will be using the most), I’ll discuss the Code book first. If you the “Open Codebook” button, a window will be opened where you can enter labels for any codes that you would like to use. In the picture below, you will see that I have assigned three labels in this example. It is also possible to assign names to the open coding tracks. To do that, first click the button “Open Coding Tracks” in the upper right of the window. We’ll not discuss the open coding interfaces here separately, because the principles that apply to these interfaces are largely the same as those for the categorical coding interfaces.

Codes

Closing the window will bring you back to the main window (which will remain there in the background if you open other windows). Let’s open the window with code descriptions next. The labels that I just assigned in the code book are now also visible in the code descriptions window. Here you can record descriptions of your codes to indicate what it is that they refer to.

Pic3

Let’s open the incident entry form next. This form has the interface for recording incident descriptions, as well as the interface for assigning codes to the descriptions. An incident description also requires you to report a date (it is possible to indicate the accuracy of the date with a drop-down menu). The code labels that you assigned in the coding book are also visible in the coding interface. You can use the tick boxes to assign the corresponding code to the incident description that is currently visible. It is also possible to write a memo for each separate incident description. On the bottom right of the screen, there is a field for working notes. These notes will remain visible if you navigate to other incidents. This can be useful if you quickly want to memorize some information while you are working on your dataset.

You’ll see a number of controls in the bottom of this screen. Some of them are self-explanatory, and I won’t discuss them in detail. One of the more important buttons is the “Put incidents in chronological order” button. If you enter incident descriptions, they won’t be put in chronological order until you press this button, or close the window and reopen it later. You can use the “+” button to create a new incident description. There is a save-button, but usually it is not even necessary to use this (Saving records is a standard option of Microsoft Access, but it really seems redundant, because every change you make seems to be saved instantly. I made the save button anyway just to be on the safe side). The arrow buttons can be used to navigate through the incidents that have already been recorded. It is often quicker to navigate to an incident with a specific data using the drop-down menu to the right (“Select Record With Specific Date”). There is also the option to go to a specific record by entering the ID in the “Go to record” field and then clicking somewhere else or pressing enter. The IDs are automatically assigned to ensure that each incident has a unique identifier. Search options are included as well (the binocular icon).

Incident Entry Form

There is also the possibility to access the so-called Datasheet view by clicking the corresponding button. I added this option to allow you to have an overview of several incidents at the same time. The datasheet view more or less corresponds to what you will see if the dataset is exported as an Excel file. It is also possible to access the coding interface from here by clicking the “+” sign to the left of each row.

Datasheet view

A final note

Unfortunately, I wasn’t able to design the export functions such that the code labels are automatically copied to your excel-file if your export your data (you’ll see labels like “code 1,” “code 2,” etc. instead). As a workaround, I have added an option to the main menu to export the code book as a separate excel file. After exporting the data and your code book, you can copy the codes from the latter file and paste it on the corresponding column names of the first file.