African Information Society Initiative [AISI] - Studies

Connectivity in Africa: Use, Benefits and Constraints of Electronic Communications - Ethiopia

Abebe Rorissa, National University of Lesotho
P. O. Roma 180, Lesotho, Southern Africa
E-Mail: r.abebe@nul.ls March 1998

Table of Contents

1. Introduction

1.1. The IDRC Impact Program

1.2. The UNECA/PADIS Project

1.3. Objectives of the Study

1.4. Organization of the Report

2. Implementation of the Study

2.1. Baseline Survey

2.2. Interviews

2.3. Problems encountered

3. Patterns of Use of Electronic Communications

4. Benefits and Constraints of Electronic Communications

5. Changes Brought by Full Internet Access

6. Summary, Conclusions and Recommendations

6.1. Summary of Major Findings

6.2. Conclusions

6.3. Recommendations

List of Tables

Table 1: Number of Completed Baseline & Revised Baseline Questionnaires Returned

Table 2: Number of Completed Questionnaires (on Changes Brought by Full Internet Access) Returned

Table 3: Distribution of Local(Ethiopian) and Expatriate(Foreigner) Respondents by Gender

Table 4: Distribution of Local(Ethiopian) and Expatriate(Foreigner) respondents by Level of Use

Table 5: Highest Qualification of the Respondents

Table 6: Computer Skills of the Respondents

Table 7: How Essential it is to Communicate Outside the Country

Table 8: Adequacy of Information Facilities at the Respondents' Organizations

Table 9: From Where Respondents Send/Receive Electronic Messages

Table 10: How Often Respondents Send/Receive Electronic Messages from their Office and/or Home

Table 11: How Respondents Usually Send Electronic Messages

Table 12: Who Covers Electronic Mail Cost

Table 13: How Often the Local & Expatriate Respondents Send/receive Electronic Messages

Table 14: Relationship Between Computer Skills of Respondents and How They Send Electronic Messages

Table 15: Relationship Between Computer Skills of Respondents and How Often They Send Electronic Messages

Table 16: What E-mail was Used for

Table 17: Percentage of Electronic Messages Sent Outside Africa, Within Africa and Within the Country

Table 18: Percentage of Electronic Messages Destined Outside Africa by Type of Organization of Respondents

Table 19: Percentage of Electronic Messages Destined Within Africa by Type of Organization of Respondents

Table 20: Percentage of Electronic Messages Destined Within the Country by Type of Organization of Respondents

Table 21: Improvements, in Ranked Order, Brought in the Respondents' Work by the Use of Electronic Communications

Table 22: Difficulties, in Ranked Order, Brought in the Respondents' Work by the Use of Electronic Communications

Table 23: Constraints to the Respondents for Making Effective Use of Electronic Communications

Table 24: Number of Respondents Who are Currently Using Internet, Have Full Internet Access and those Who Plan to Obtain Full Internet Access

Table 25: Reasons Given by Respondents Who do not have Plans to Obtain Full Internet Access.

Table 26: Type(s) of Internet Services that Prompts/Prompted Respondents to Subscribe

Table 27: Types of Internet Services Respodents Use and Feel Useful

Table 28: How Respondents Feel about the Adequacy of Telecommunications Facilities Available at their Organizations and in their Area for Full Internet Access

Table 29: Adequacy of Information Facilities at the Respondents’ Organizations

Table 30: Improvements, in Ranked Order, Brought in the Respondents' Work by the Use of Internet

Table 31: Technical or other Constraints Preventing Respondents from Making a more Effective Use of Internet

Table 32: Improvements, in Ranked Order, which Respondents Expect to be Brought to their Work by the Use of Internet

Table 33: Technical or other Constraints which Respondents Expect May Prevent them from Making a more Effective Use of Internet

Appendix I: Baseline Questionnaire

Appendix II: Revised Baseline Questionnaire.

Appendix III: Procedures Used for Classifying PADISnet’s E-Mail

Entry Points and for Selecting a Sample of 50 Points

for interview.

Appendix IV: Questionnaire on Changes Brought by Full Internet Access

Appendix V: Interview Schedule

1. Introduction

1.1. The IDRC Impact Program

In 1992, the International Development Research Center (IDRC) of Canada organized an exploratory workshop in which researchers from almost all the continents participated in. The main theme of the workshop was the impact of information on development.

An electronic conference, on top of the above mentioned workshop, gave rise to the production, in 1993, of an important publication in the area of impact assessment, "Measuring the Impact of Information on Development", edited by Michel Menou and published by the IDRC.

The entire IDRC Impact Program was divided into two phases. The first phase identified key issues and formulated "a preliminary framework for future investigations" while the second phase of the IDRC Impact Program, which run from 1994 to 1998, focused on testing "the preliminary framework and get hands-on experience of the problems associated with this unusual kind of investigation." A total of eight (8) projects were launched during the second phase to test the framework under different situations.

1.2. The UNECA/PADIS Project

The IDRC sponsored project "Capacity Building for Electronic Communication in Africa" (CABECA), which was implemented by the Pan African Development Information System (PADIS) of the United Nations Economic Commission for Africa (UNECA) introduced low cost electronic communication networks in many African countries since 1993.

The UNECA/PADIS project which is one of the eight IDRC supported projects mentioned above (1.1) and being reported here involved four of the countries (Ethiopia, Senegal, Uganda and Zambia), which were beneficiaries of the CABECA project. It was originally entitled "Impact of Electronic Communications on Development in Africa", consonant with the overall theme of the program of the International Development Research Centre (IDRC), Canada, through which it was supported. The research team felt that this title sounded too broad and ambitious with respect to the actual scope and outcome of the project. They thus decided to use the above main title for the final reports of the project.

The final report of the project is made of a series of self-contained documents with a view to facilitate their further dissemination and use. The series includes:

Extended abstracts of the synthesis report and national reports;
A synthesis report in two parts, one dealing with methodological issues, the other one with the overall findings and conclusions of the project;
The reports of phase one and phase two of the national studies in Senegal, Uganda and Zambia and a consolidated report for phase one and two in Ethiopia, all in a revised version.

The reports will be available in print or electronic form from the Pan African Development Information System (PADIS) of the United Nations Economic Commission for Africa. It is also planned to post, at a later stage, an eventually adapted version of all of them on the Web site of the African Information Society Initiative (http://www.uneca.org/aisi).

The present document is devoted to reporting results of a two-phase survey of users of the PADISnet FidoNet node in Addis Ababa, Ethiopia.

1.3. Objectives of the Study

The main objectives of the UNECA/PADIS study were "to contribute to the development of indicators for the measurement of the impact of information and to assess the impact of electronic communications (telematics) on development in Africa."

However, due to various reasons (constraints in time and other resources), the study failed to achieve its objectives.

1.4. Organization of the Report

This report has six sections. The first section outlines the IDRC Impact Program and the UNECA/PADIS Project together with its objectives. Section two deals with detailed procedures followed in the execution of the study. In sections three and four, patterns of use of electronic communications and their benefits and constraints, respectively, as indicated by users of the PADISnet through their responses to the baseline and revised baseline questionnaires (Appendix I and II) and interviews (Appendix V) are presented. Results obtained from the questionnaire on changes brought by full Internet access (Appendix IV) are presented under section five. The last section gives summary of major findings, conclusions and recommendations for future studies.

2. Implementation of the Study

2.1. Baseline Survey

No.	Activity	Duration/Date
1	Identification of entry points of the PADISnet and their classification into five(5) groups, namely Private/Individual, Academic & research, International, Governmental, and Non-governmental	March 1995
2	Identification of active entry points during the month of October 1994 (222 entry points were identified)	March 1995
3	Compilation of traffic data (for October 1994) of all the active entry points during the month of October 1994	April 1995
4	Distribution of a separate questionnaire prepared for participants of the April 1995 Regional Symposium on Telematics for Development in Africa, organized by PADIS	April 1995
5	Since the traffic data of a single month was not found sufficient for classification of the entry points into various levels of use (low, medium, high and very high), identification of all active entry points during January and February 1995 was necessary and a total of 215 entry points were identified	June 1995
6	Exclusion of entry points being used by embassies and users outside the country was made and this decreased the number of active entry points during January & February 1995 to 206 (see Appendix III for the distribution of the 206 points by level of use and type of organization)	June 1995
7	Classification of the active entry points into low, medium, high and very high levels of use and working out the sample structure of the entry points for interviews (see Appendix III)	August 1995
8	Distribution (via E-mail) of the baseline questionnaire (Appendix I) to all the 206 entry points identified as active during January & February 1995	August 1995
9	Posting of the sample structure onto the INIMCAS-L Listserv	August 1995
10	Identification and selection of 50 entry points for interviews	September 1995

No.	Activity	Duration/Date
11	Collection of the distributed questionnaires	August - October, 1995
12	Redistribution and collection (via E-mail and by handing out) of the baseline questionnaire (2nd round) (see Table 1 below for the number of returned questionnaires)	November - December 1995
13	Interviewing the sample of entry points (interview questions were first sent to them via E-mail to see how they would react to them)	December 1995 - March 1996
14	Writing up of the Phase I report	May 1996
15	Distribution (by E-Mail) of the revised baseline questionnaire (Appendix II) and the questionnaire on changes brought by full Internet access. A combined questionnaire(of these two) was sent to those entry points from which no response was received during the 1^st phase while the one on changes brought by full Internet access was sent to all the 206 entry points of the PADISnet identified as active during January & February 1995	October 1996
16	Since, among the users to which the questionnaires were sent via E-Mail, only two have responded, the two sets of questionnaires had to be handed out to the respondents and appointments to collect them had to be fixed	March -May 1997
17	Interviews were conducted with additional users from the sample selected during the 1st phase (to make the total number of interviewees 50) and to check whether those responses received during the 1st phase still hold	March 1997
18	Writing up of the Phase II report	August 1997
19	Writing up of the final report	March 1998

Table 1: Number of Completed Baseline and Revised Baseline Questionnaires Returned

Level of Use*	Type of Organization
Level of Use*	P	A	I	G	NG	Total	% of Total**
Low	11	14	15	2	9	51	50.5
Medium	3	9	-	-	3	15	14.8
High	1	5	3	-	3	12	11.9
Very High	3	12	6	1	1	23	22.8
Total	18	40	24	3	16	101	100
% of Total	17.8	39.6	23.8	3.0	15.8	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-governmental, *This classification was made based on PADISnet’s traffic data for January & February 1995; **N=101

Table 1 above shows that users affiliated to Academic and Research institutions and users from the low level of use group were the majority of the respondents. (see Appendix III for the original structure of the 206 entry points)

Table 2: Number of Completed Questionnaires on Changes Brought by Full Internet Access Returned

Level of Use*	Type of Organization
Level of Use*	P	A	I	G	NG	Total	% of Total**
Low	16	10	8	3	2	39	67.2
Medium	1	3	-	-	1	5	8.6
High	-	1	1	-	1	3	5.2
Very High	2	6	1	1	1	11	19.0
Total	19	20	10	4	5	58	100
% of Total**	32.8	34.5	17.2	6.9	8.6	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-governmental *This classification was made based on PADISnet’s traffic data for January & February 1995; **N=58

According to Table 2 above, private/individual users of the PADISnet and users affiliated to Academic and Research institutions as well as users from the low level of use group were the majority of the respondents.

2.2. Interviews

Interviewing was the most difficult as well as least productive part of the study. A total of 33 out of the original sample of 50 users were interviewed. Out of the 33 users interviewed, 22 were interviewed between December 1995 and March 1996 while the remaining 11 were interviewed during the month of March 1997.

The organizational affiliations of these 33 users are as follows: 11 were from private/individual, 8 were from academic & research, 7 were from non-governmental, 5 were from international and 2 were from governmental organizations.

2.3. Problems encountered

1. There were much less number of respondents, than expected, to the baseline questionnaire and, as a result, it was difficult to interview a satisfactory proportion of the target of 50 entry points.

2. What have been identified as active entry points at the beginning of the survey were later discovered to be inactive due to various reasons (such as broken machine or modem, people being on long vacations and away from office or leaving the country or organization, etc.)

3. Unavailability of some users for interview sessions demanded for repeated visits to a single user who, in some cases, gives very little information when interviewed. For instance, a lecturer said that his problem is related to getting a chalk rather than communication related.

4. As some of the former PADISnet users have already been using the Internet services (being provided by the Ethiopian Telecommunications Corporation) for electronic communications and much more, and since they have either not been using their PADISnet entry points or they have been using it rarely, they were very reluctant to fill in the questionnaires and gave various excuses when asked to be available for interview sessions. Therefore, this resulted, once again, in low response rate.

5. Some of the respondents didn’t respond to the open-ended questions and some of them didn’t list their responses in ranked order, where they were asked to do so. As a result, analysis of results was difficult.

3. Patterns of Use of Electronic Communications

Out of 25 respondents who indicated that they are the sole users of their addresses, about half (12=48%) are private/individual users with 5, 4, 3 and 1 respondents from academic & research, non-governmental, international and governmental organizations, respectively. Moreover, 15(60%) of the sole users were expatriates.

Table 3: Distribution of Local(Ethiopian) and Expatriate(Foreigner) respondents by Gender

	Type of Organization
	P	A	I	G	NG	Total	% of Total
Local:	4	36	8	3	8	59	58.4*
Male	3	34	8	2	6	53	89.8**
Female	1	2	-	1	2	6	10.2**
Expatriate:	14	4	16	-	8	42	41.6*
Male	13	3	11	-	7	34	81.0***
Female	1	1	5	-	1	8	19.0***
Permanent Residents	6	2	2	-	3	13	31.0***
Temporary Residents	8	2	14	-	5	29	69.0***

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-Governmental, *N=101, **N=59, ***N=42

Female respondents were very few (13.9%) compared to male respondents (86.1%). Respondents from academic & research institutions are also the majority (61.0%) of the local users while 71.4% of the expatriates were private/individual users and users from International organizations. Out of the 42 expatriate respondents, more than two third were temporary residents (i.e. less than 5 years).

Table 4: Distribution of Local(Ethiopian) and Expatriate(Foreigner) respondents by Level of Use

Local/Expatriate	Level of Use
Local/Expatriate	Low	Medium	High	Very High	Total
Local:	32	11	4	12	59
Male	27	10	4	12	53
Female	5	1	-	-	6
Expatriate:	19	4	8	11	42
Male	17	4	7	6	34
Female	2	-	1	5	8

Table 4 above shows that the majority of both local and expatriate respondents were from the low and very high levels of use groups. This is due to the fact that the low and very high level of use group of entry points were the 1^st and the 2^nd largest groups of all the PADISnet users.

Table 5: Highest Qualification of the Respondents

Local or Expatriate	Highest Qualification
Local or Expatriate	High Sch. Cert.	Diploma	BA/BSc	MA/MSc	PhD/MD	Total	% of Total*
Local	1	3	24	17	14	59	59.0
Expatriate	-	-	11	12	18	41	41.0
Total	1	3	35	29	32	100	100
% of Total*	1.0	3.0	35.0	29.0	32.0	100

*N=100

Almost all the respondents (96.0%) are holders of BA/BSc degree or above. On the average, the higher the qualification the greater the number of respondents.

Table 6: Computer Skills of the Respondents

Computer Skill	Highest Qualification
Computer Skill	High Sch. Cert.	Diploma	BA/BSc	MA/MSc	PhD/MD	Total	% of Total*
Experienced	1	3	26	15	15	60	60.0
Intermediate	-	-	8	12	17	37	37.0
Beginner	-	-	1	2	-	3	3.0
Total	1	3	35	29	32	100	100
% of Total*	1.0	3.0	35.0	29.0	32.0	100

*N=100

Qualifications and computer skills of the respondents do not seem to be positively correlated for experienced computer users though they do for those who rated their computer skills as intermediate. Only 3.0% of the respondents are beginners while the rest 97.0% are intermediate and experienced computer users, the majority of them (60%) being experienced computer users.

Table 7: How Essential it is to Communicate Outside the Country

How Essential	Type of Organization
How Essential	P	A	I	G	NG	Total	% of Total*
Essential	15	36	24	3	15	93	93.9
Sometimes Essential	1	3	-	-	1	5	5.1
Seldom Essential	-	1	-	-	-	1	1.0
Total	16	40	24	3	16	99	100
% of Total*	16.2	40.4	24.2	3.0	16.2	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-Governmental, *N=99

As far as essentiality of communicating outside the Country is concerned, 93.9% of them indicated that it is essential while only 1.0% of them think that it is seldom essential.

Table 8: Adequacy of Information Facilities at the Respondents' Organizations

Adequacy	Type of Organization
Adequacy	P	A	I	G	NG	Total	% of Total*
Adequate	4	7	12	2	6	31	31.3
Inadequate	8	33	11	1	7	60	60.6
Not Available	4	-	1	-	3	8	8.1
Total	16	40	24	3	16	99	100
% of Total*	16.2	40.4	24.2	3.0	16.2	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-Governmental, *N=99

Except respondents from international and governmental organizations, the majority of those from other types of organizations feel that information facilities at their organizations are inadequate. Moreover, respondents from academic & research institutions, who heavily rely on libraries and information centers for their day to day activities, feel more so. Their opinion conforms with the usual criticism about the inadequacy, in terms of quality as well as quantity of information, of academic and research libraries in the country.

Table 9: From Where Respondents Send/Receive Electronic Messages

From Where They Send/Receive E-Mail	Type of Organization
From Where They Send/Receive E-Mail	P	A	I	G	NG	Total	% of Total*
Office	5	37	19	3	14	78	77.2
Home	6	-	1	-	-	7	7.0
Both Office & Home	7	3	4	-	2	16	15.8
Total	18	40	24	3	16	101	100
% of Total*	17.8	39.6	23.8	3.0	15.8	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-Governmental, *N=101

Tables 9 and 10 show that home computers in general and E-mail facilities at home in particular are not common among the respondents and that is why more than three fourth of them indicated that they send/receive electronic messages only from their offices, with the exception of about half of the private/individual users who send/receive electronic messages both from their offices and homes. It is also apparent from Table 10 that the majority (75%) of the respondents send/receive electronic messages at least once per week.

Table 10: How Often Respondents Send/Receive Electronic Messages from their Office and/or Home

How Often they Send/receive E-Mail	Where they Send/receive E-Mail from
	Office	Home	Both Office & Home	Total	% of Total*
Less than once per month	8	-	2	10	10.1
1 to 4 times per month	12	1	2	15	15.1
1 to 5 times per week	32	3	2	37	37.4
More than once per day	25	3	9	37	37.4
Total	77	7	15	99	100
% of Total*	77.8	7.1	15.1	100

*N=99

Table 11: How Respondents Usually Send Electronic Messages

How They Send E-mail	Type of Organization
How They Send E-mail	P	A	I	G	NG	Total	% of Total*
Themselves	17	38	21	3	11	90	90.0
Through someone else	1	2	2	-	5	10	10.0
Total	18	40	23	3	16	100	100
% of Total*	18.0	40.0	23.0	3.0	16.0	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-Governmental, *N=100

It is clear from Table 11 above that most (90%) of the respondents send electronic messages by logging in and typing for themselves. This conforms to the fact that E-Mail is more personalized than, for instance, fax.

Out of a total of 21 respondents who subscribe to other networks (on top of PADISnet), about half (47.6%) of them are affiliated to international organizations which have electronic connections to their respective head offices abroad. The Ethiopian Telecommunications Authority’s network (which provides full Internet services), UN CC-Mail(?) and HealthNet are the networks, other than PADISnet, used by the 21 respondents.

Table 12: Who Covers Electronic Mail Cost

E-mail Cost is Covered by	Type of Organization
E-mail Cost is Covered by	P	A	I	G	NG	Total	% of Total*
Yourself	9	-	4	-	1	14	14.0
Your Organization	2	30	10	2	2	46	46.0
A Project/Other	7	9	10	1	13	40	40.0
Total	18	39	24	3	16	100	100
% of Total*	18.0	39.0	24.0	3.0	16.0	100

P=Private/Individual, A=Academic & Research, I=International, G=Governmental, NG=Non-Governmental, *N=100

Electronic mail cost is mostly covered by either the respondents’ organizations or projects. Only 14.0% of the respondents cover the cost for themselves, among which (64.3%) are private/individual users.

Table 13: How Often the Local & Expatriate Respondents Send/receive Electronic Messages

How Often they Send/receive E-Mail	Local/Expatriate
	Local	Expatriate	Total	% of Total*
Less than once per month	8	2	10	10.1
1 to 4 times per month	9	6	15	15.1
1 to 5 times per week	23	14	37	37.4
More than once per day	17	20	37	37.4
Total	57	42	99	100
% of Total*	57.6	42.4	100

*N=99

According to Table 13 above, there is no difference between the locals & expatriates in the frequency with which they send electronic messages. More than three fourth of the respondents send an average of one message per working day (Monday - Friday).

Table 14: Relationship Between Computer Skills of Respondents and How They Send Electronic Messages

How They Send E-mail	Computer Skills
How They Send E-mail	Experienced	Intermediate	Beginner	Total	% of Total*
Themselves	57	32	1	90	90.9
Through Someone else	3	4	2	9	9.1
<