The Interview And Other Data Gathering Methods

The interview is the primary technique for information gathering during the systems analysis phases of a development project. It is a skill which must be mastered by every analyst. The interviewing skills of the analyst determine what information is gathered, and the quality and depth of that information. Interviewing, observation, and research are the primary tools of the analyst.

This chapter provides a discussion of the interview and its importance, interview guidelines, and guidelines on interview documentation.

What Is an Interview?

A definition

An interview is "a formal face-to-face meeting, especially, one arranged for the assessment of the qualifications of an applicant, as for employment or admission.... A conversation, as one conducted by a reporter, in which facts, or statements are elicited from another." (The American Heritage Dictionary, Second College Edition)

The interview is the primary technique for information gathering during the systems analysis phases of a development project. It is a skill which must be mastered by every analyst. The interviewing skills of the analyst determine what information is gathered, and the quality and depth of that information. Interviewing, observation, and research are the primary tools of the analyst.

The interview is a specific form of meeting or conference, and is usually limited to two persons, the interviewer and the interviewee. In special circumstances there may be more than one interviewer or more than one interviewee in attendance. In these cases there should still be one primary interviewer and one primary interviewee.

Types of Interviews

During the analysis process, interviews are conducted for a variety of purposes and with a variety of goals in mind. An interview can be conducted at various times within the process for

1. Initial introduction
2. Familiarization or background
3. Fact gathering
4. Verification of information gathered elsewhere
5. Confirmation of information gathered from the interviewee
6. Follow-up, amplification, and clarification

Interviewing Components

The interview process itself consists of a number of parts.

1. Selection of the interviewee and scheduling time for the interview
2. Preparation of interview questions, or script
3. The interview itself
4. Documentation of the facts and information gathered during the interview
5. Review of the interview write up with the interviewee
6. Correction of the write up, sign-off, and filing

What Are the Goals of the Interview?

At each level, each phase, and with each interviewee, an interview may be conducted to:

1. Gather information on the company
2. Gather information on the function
3. Gather information on processes or activities
4. Uncover problems
5. Conduct a needs determination
6. Verification of previously gathered facts
7. Gather opinions or viewpoints
8. Provide information
9. Obtain leads for further interviews

Interviewing Guidelines

Given these various phases and the variety of goals of an interview, the importance of a properly conducted interview should be self-evident. Since each interview is in fact a personal exchange of information between two personalities, a set of guidelines for the interviewer should be established to ensure that nothing interferes with the stated goal, i.e., gathering complete, accurate information. The interview is not an adversary relationship; instead it should be a conversation. Above all it is a process, and like most processes it has certain rules and guidelines which should be followed.

1. First and foremost, establish the tone of the interview.
2. Let the interviewee know the reason for the interview and why he or she was selected to be interviewed.
3. Stress that the interviewee's knowledge and opinions are important, and will aid in the analysis process.
4. Gain the interviewee's trust and cooperation early on, and maintain it throughout.
5. Establish what will happen to the information gathered.
6. Determine any areas of confidentiality or restricted information.
7. Let the interviewee know that candor and honesty will be valued and that nothing will be published or passed on until it has been reviewed and verified by the interviewee.
8. Firmly establish that there are no negative consequences to being interviewed.

Dos and Don'ts of Interviewing

The rules of interviewing are similar to the rules which govern most human interactions and to the rules which govern most investigative and problem-solving processes. In effect they can be called the rules of the game.

1. Do not assume anything.
2. Do not form pre-judgments.
3. Do ask questions which start with who, what, where, when, why, and how, where possible.
4. Do ask both open and closed questions.
5. Do verify understanding through probing and confirming questions.
6. Do avoid confrontation.
7. Do act in a friendly but professional manner.
8. Do not interrupt.
9. Do listen actively.
10. Do take notes, but do not be obtrusive about it.
11. Do let the interviewee do most of the talking
12. Do establish rapport early and maintain it.
13. Do maintain control over the subject matter.
14. Do not go off on tangents.
15. Do establish a time frame for the interview and stick to it.
16. Do conclude positively.
17. Do allow for follow-up or clarification interviews later on.
18. Do be polite and courteous.

Who to Interview

One of the analyst's first and most important tasks during the data gathering phase of the analysis process, is to determine who has to be interviewed. This includes selecting the interviewees, understanding what can be expected from an interview of a person at a specific level, how to verify the information received from an interview, and, most important, understanding the perspective of the person being interviewed.

Most analysis projects have a user liaison assigned to the analysis team. It is this person's function to introduce the analyst to those being interviewed, to provide background information, and to interpret (or translate) the information which is obtained from the interviews. This person usually has the additional role of assisting the analyst in choosing those to be interviewed, scheduling the interviews themselves, and in some cases attending the interviews.

Under normal conditions, the analyst will have access to all people in the user area, although normally there is no need to interview them all. This is especially true if the user area is very large.

Generally speaking, the list of those to be interviewed can be divided into three sections:

1. the most senior manager,
2. his or her subordinates and junior managers, and
3. line workers, clerks, production people, sales staff, etc. See Figure 7.1 for interviewing and verification sequences.

The following are some guidelines for the analyst as to who to interview, when to interview them, what their perspective is, and what to expect from the interview (the goal of the interview itself).

1. The most senior manager in the user area. It is vital to interview this person at the start of the project. From him or her the analyst will obtain an overview of the user area, an overview of the functions performed by that area, and an idea of how the area fits within the overall structure of the organization and its activities.

   The analyst can expect that the information received here will be general in nature. However, the manager will be able to define fairly accurately the business objectives of the project, the functions for which support is needed, any perceived problems which require attention, the time frame within which the project is expected to be completed, and the constraints, both business and financial, which apply. This manager can also suggest persons to interview, areas to concentrate on, and other sources of information. This manager should also be able to provide the enterprise perspective which is vital to the analyst's understanding.

   For each area within his or her control, the manager should be able to give the analyst an organization chart which indicates the structure of the area, the number of people within it, and an overview of its function. These charts should be used to select the names of individuals to be interviewed: the area manager, interviewees recommended by the manager, and, if necessary, alternative contacts in each area.

   The analyst should not expect great levels of detail about the individual activities of the area nor about the individual tasks which are performed, much less how they are performed.

   Regardless of any other information acquired, it is this senior manager who will benefit most from the project, who in most cases is funding the project itself, and who will in all probability have the final sign-off on the completed work. It is vital, therefore, to have a clear undestanding of what is expected by this person.

2. Immediate subordinates of the most senior manager and junior managers. There may be many levels of these managers, each lower one having a smaller span of control, more specific responsibilities, and less authority than the one immediately above. Working from the organization charts supplied by the senior area manager, the analyst should schedule meetings with either (a) each of the senior manager's immediate subordinates or (b) those subordinates for whom the project is being undertaken. In either case (a) or (b), it may be necessary for the analyst to speak to each of these managers, if only to arrange to speak to people in their areas who might be affected by the project or who might be able to contribute to the information gathering process of the analysis.

   It may also be found that the actual scope of the project, or the problem, is larger than originally defined and that the sources of the problem are in areas other than those which are experiencing the symptoms. For this reason it is usually a good idea to request and receive explicit permission from the senior manager to interview each of the second-line managers.

   The senior manager should also be asked to explain to them the scope of the project, its intent, and that it may be necessary to interview some of their subordinates as the project progresses even though they are not directly involved (initially) in the project.

   In many cases, people move from function to function, or at least from activity to activity during their employment by the firm. Many of these people may have more knowledge about the area under analysis than the current incumbents.

   In many cases the managers themselves will have been promoted through the ranks, and will retain, if not current information about the area activities, at least some of the historical perspective as to how and why the area performs some of its activities and how they have changed over the years. In addition they can provide additional detail as to any problems, the reasons for those problems, and suggested ways to resolve them.

   The managers at the various levels may also be able to clarify the statements made by their immediate superiors. The analyst will find that each person has a different perspective on an area, a perspective which reflects the individual's responsibilities and authorities. Each individual will also be able to provide a more detailed perspective on the interactions between different areas.

   One of the things the analyst should be doing constantly during these interviews is verifying and cross-checking the information received. This cross-checking should be done in an objective manner and should not violate any information given in confidence. The objective is not to place blame or point fingers but to ensure that the information is correct. Where discrepancies arise between the information received from different managers, it may be necessary to verify the original information or seek a third source.

3. Operational and clerical personnel. These people actually perform the detailed tasks of processing, manufacturing, data gathering, or reporting. It is their tasks that the automated systems are ultimately intended to perform or augment. For this reason it is mandatory that the analyst thoroughly understand not only what they are supposed to do, but what they actually do, how they do it, and why they do it. It can be expected that most interviews will occur at this level. They will also be the most detailed.

   Generally, the analyst will want to interview at least one person from each specific task or activity area. In some cases it may be necessary to interview more than one person. Each person interviewed will in all probability refer to something received from or passed to another person or area. The analyst must verify that both individuals agree on the identity of the materials they are passing to each other and that both agree on the content of the transferred material. The understanding of the operations level personnel as to their specific duties, functions, and activities must match that of their immediate managers. Any discrepancies in this area must be resolved by the analyst.

The Need for Documentation

Everyone talks about the weather but no one can do anything about it. In the case of documentation, everyone talks about it but few do it; however, unlike the weather, most people can document, and document effectively.

Documentation, however painful and tedious it may seem, is one of the most critical tasks of analysis. The documentation produced as a result of the analytical interviews, the analyst's observations and research, and ultimately, the total analysis phase of the project serves a number of purposes.

Permanence. The need for documentation is rooted in semantics and human memory. Verbal communications are both transitory and subject to interpretation. The average person has a language working set of about 500 to 1000 words. The written working set, by contrast, is much larger, perhaps by as much as an order of magnitude. Verbal communication is augmented by inflection, body language, and by a process of feedback and interaction, all of which serve to clarify the ambiguous, the ill-defined, and ill-understood. Human memory is imperfect. Words communicated verbally can only be recalled and examined with difficulty, if at all.

Precision and recall. A written document is more precise and may be reviewed repeatedly until understanding is achieved. It has the added advantage that small changes can be made to it without having to restate the entire premise or thought. Additionally, once an idea is written down, it may be recalled at will exactly as first presented and may be completed by someone other than the original author, or authors. Because there is little feedback from the written word, one can only take issue with misstatements of fact or with ambiguous wording. If it isn't written down, it isn't there.

Graphics. Documentation usually includes both a narrative portion and an illustration portion. These illustrations serve to amplify and enhance the narrative. One picture can be worth many thousands of words, if properly drawn. The graphics of the analytical documentation, whether it employs simple flowcharts, HIPO diagrams, data flow diagrams, or powerful modeling techniques such as those based upon the entity-relationship-attribute approach, presents the user and the analyst with a way to walk through the picture developed from the analysis, and ultimately walk through the design developed from the analysis. These walk-through sessions enable both to understand the environment and to detect any ambiguities and anomalies. Illustrations have the added advantage that they can be viewed in their entirety, whereas narrative may only be viewed in fragments.

Functions of Documentation

Documentation serves to clarify understanding, and perhaps most important, it provides the audit trail of the analyst. That is, it creates the records which can be referred to at some later date and which serve as the basis for future work and decisions.

Good documentation precludes the need to return to the interviewee for a repetition of ground previously covered. Good documentation can be reviewed over and over until adequate understanding is achieved.

Documentation is tedious and sometimes boring. But it is also vital. Good documentation allows other analysts and the analyst's successors to pick up where the first left off, should he or she be reassigned. Documentation is necessary if the next project phases are to be successful, since they are predicated on the results of the analysis. To a very real extent, analytical documentation provides the road maps for the remainder of the project. If the maps are faulty, or incomplete, the succeeding teams may wind up in a swamp, or worse, in quicksand.

Most important, the finalized documentation serves as a contract between the user and the data processing developer. In it the analyst has described the user's environment, the analyst's understanding of the user's needs and requirements, and with the proposal for a future environment, the analyst's description of the system to be designed and built by the developers. With the user's sign-off, or approval of these documents, a contract is created between the two. Barring unforeseen changes in the business environment, the problems described in the documentation will be rectified and the environment proposed will be the one built and installed for the user.

The document becomes, in effect, a statement of the work to be performed. The time to modify and change it is before the work begins; afterward it may be too late. From the developer's perspective, any post sign-off changes may require a re-negotiation of either time frames, costs, or resources. From the user's perspective, the design is what is contracted for and what he or she is paying for. If the final product does not conform to the proposal, then it is up to the developer to rework the product until the user is satisfied.

Documenting the Interview

The interview is not complete until it has been documented. See Figure 7.2 for a header form for sample interview notes. The documentation of the interview need not be a verbatim transcript of what was said but should cover the following items.

1. Who was interviewed and who did the interviewing, including the title of the interviewee, the interviewee's function and immediate superior, and interviewer name and title
2. The date, time, and location where the interview took place
3. Names and titles of any other persons who were in attendance
4. The stated objective of the interview
5. The interviewee's job or functional description
6. The interviewee's organizational chart and organizational charter, if appropriate. A complete description of any facts which were obtained during the interview
7. A complete description of any opinions stated by those interviewed
8. Any conclusions drawn from the facts or opinions as presented
9. Any business problems uncovered during the interview
10. Samples of any forms, reports, charts, graphs, documents, manuals, policies, standards, or procedures referred to or discussed during the interview
11. Any charts or diagrams drawn as a result of the information gathered during the interview
12. Any relevant comments made by the interviewee
13. Any numbers, transaction or document volume information, task timing information, capacity information, quality information, etc., gathered during the interview

Dos and Don'ts of Documentation

1. Keep a clean separation between fact and opinion, whether the opinions are those of the user or those of the analyst. In your documentation, be sure to identify clearly which parts are fact and which are opinion.
2. Document all sources of information, especially all facts and numbers.
3. Make sure that all documentation is dated, not only with the date of the original version, but also with the dates of any revisions. If documentation has been revised, a revision identification scheme should identify which revision is most current. See Figure 7.3.
4. Ensure that all pages of the documentation are numbered and that there is a table of contents which indicates where each major section may be located.
5. Analytical documentation tends to be long, sometimes overly long; however, it is better to err on the side of length than to leave out information which may become important later on.
6. Write for your audience, which implies that your audience should be known. Technical information should be written for technical audiences; user documentation should be written for users.
7. All documentation should contain an executive summary which contains salient background material, findings, conclusions, and recommendations, in that order. This executive summary material should be explained in depth in the support materials. See Figure 7.4 for a major section table of contents as it would appear in the documentation.
8. Studies have shown that most people will only be interested in the first 20 to 30 pages of any report. All the information you need to convey should be in those first 20 to 30 pages; any back-up material should be reserved for appendices and supporting documentation.
9. One picture is worth a thousand words. Pictures could include graphs, charts, tables, and diagrams. They are highly informative and easily understood. Use them liberally. The documentation should contain a listing of the location of all graphs, charts, and diagrams. Remember. When using charts, diagrams, etc., that all special symbols should be identified in a legend, along with the source of the information for the charts, diagrams, etc. All axes of charts and graphs should be clearly labeled. Chart and graph sealing information should be clearly in evidence. All pictures should be clearly titled.
10. Acronyms may be used, but they should be spelled out in full at their first use. The documentation should contain a list of these acronyms along with their full word meanings. If in doubt, spell it out.
11. All businesses and functions have their special terminology. Its use is unavoidable, and in fact, it would be unwise not to use it. All documentation should contain a glossary which defines these special terms.
12. The best analysis can be discredited by sloppy documentation. The documentation doesn't have to be professionally done, but it should be neatly typed, with no spelling errors or typos. The grammar should be reasonably correct. The final document should be neatly bound with cover pages.
13. Avoid verbosity. Avoid pomposity. The document should be written in business English with reasonably short sentences. This is not the place to try and impress people with the extent of your vocabulary. Use headings and subheadings liberally.
14. Most large documents will have sections which are written by more than one person. Try to ensure some level of stylistic consistency. The document should flow and should be easily readable. Remember it is meant to convey information.

Report Format Style Guidelines

The appearance and readability of documentation can be greatly enhanced by following these guidelines.

Use a consistent outline format that makes it easy for the reader to determine the current location the document structure.  Although mixed numeric and alpha outlining is frequently used, in can be very confusing, especially if the full item parentage is not used. Numeric outlining, on the other hand makes it easy for the reader to determine current location. For example,

1.0
1.1
1.1.1
1.1.1.1
1.1.1.1.1

If it becomes necessary to itemize under any heading use lower case roman numerals.

i)
ii)
iii)
iv)

An indentation should always have more than one item.

Every outlined section should always have a boldfaced heading.

Do not start a paragraph with a number or a single letter.

Avoid the use of bullets or special marks in final reports--they give the impression that the report is still a summary

Make sure that all words are properly spelled and that proper grammar is used. Nothing can be more detrimental to a report than improper grammar or word usage. Avoid slang, technical jargon, and idiom where possible.

The following are general stylistic considerations which apply.

1. Do not use the term "etc." or "et al." in a report. If the other items are known list them. If the list is too long to include in the text, or if the items are relatively minor, put them in an appendix.
2. Do not use the terms "same as" or "ditto" in the final text. Always use sentences and paragraphs, and avoid freestanding phrases.
3. All exhibits should be meaningfully titled and referred to in the text. Exhibits should follow, or be as close as possible to, the page of reference.
4. All financial numbers should be checked for arithmetic. That is they should cross and down-foot.
5. All exhibits should be consistent from exhibit to exhibit, from exhibit to text, and chapter to chapter. Be sure to use consistent units of measure and time frames in all exhibits.
6. When numbers are used within the text, make sure that the units of measure are indicated clearly. Single number references should be spelled out in words and followed by their numeric equivalent, e.g., six (6), thirteen (13), twenty-two (22). All monetary figures should be preceded by the appropriate currency symbol and show all commas and decimal points for clarity (i.e., $1,000.00)

Non-Interview Data Gathering Methods

In today’s business environment, when almost all office workers are provided with a personal computer, and almost all managers have one as well, it has become increasingly necessary to extend the scope of the data gathering process to a larger portion of the workforce, in some cases it has become necessary to gather data from all employees in a given areas, or even in extreme cases throughout the firm.  Obviously in these cases interviewing everyone is impractical.

There are several data gathering methods which are used in these cases:

Phone Interviews – These are used to gather non-sensitive yes or no types of answers to highly focused questions. 

Characteristics:

1. Sample size is relatively limited.
2. The response rate is relatively good
3. The turnaround time is very fast
4. moderate cost
5. Good for gathering relatively simple data

Table 7.1  Advantages and Disadvantages of Phone Interviews

Mail Surveys - These are used to gather broad quantifiable data from a large sample.

Characteristics: