DATA COLLECTION


The problem solving method is about taking a structured approach to understanding a client's predicament. We saw, however, that there are some areas where you can confidently assume the correctness of your views while in others they should be regarded as hypothetical or speculative.

Figure 6.7 shows the stages involved. Data collection is shown as a means by which you can test hypotheses, proving them to be true or false or at least shedding a little light on them. But we should never lose sight of the fact that consultancy projects are a means to an end, that end being to the benefit of the client. So, the figure can be redrawn as an input-output model, as shown in Figure 6.9. The scope of the consultancy project will determine the nature of the deliverables to the client. Hypotheses can be regarded as provisional conclusions, while conclusions can be regarded as proven hypotheses. The data that is collected will be a function of the data specification. So there is a symmetry about inputs and outputs as well as the sequence shown in Figure 6.7. Data collection is key to this process.

click to expand
Figure 6.9: The problem solving approach as an input-output model

There are two essential considerations in data collection: defining what data is required; and using an appropriate method to gather it.

Defining What Data is Required

Data collection is a time-consuming element in any consultancy project and so one of the most expensive. It is essential that you are clear about what data you are aiming to collect and why you are doing so. It is very easy for the intellectually curious to go along some cul-de-sacs and expensively collect data that has little relevance to the questions in hand. The process of data collection should therefore be directed to verifying the hypotheses that you have selected.

On this basis, you could produce a specification of what data is required, and then go out and collect it. In practice, consultancy requires more. One feature of first engaging with a new client or assignment is the process of familiarization. The consultant has to get to know the client - the people, the culture, the business processes, etc - and the client has to learn how to accommodate the consultant and the project being undertaken. Each is progressing along a learning curve about the other, and it is unusual for familiarization to be embodied in a formal process of data collection.

Figure 6.10 shows how the scope of a project evolves and the nature of data collection that goes along with it. Early in the project, you need to develop a feel for the political environment in the client. Data collection therefore is open-ended, and you will be concerned with opinions as much as facts. Next, you will generate and revise hypotheses during this scoping and familiarization phase (it is worth noting that you may wish to revise your hypotheses as a result of data collection). Once you have established your shortlist of hypotheses, then your data collection can become more focused on the data you need to check out your selected hypotheses.

click to expand
Figure 6.10: Data collection evolves with the project

Data collection can be of two broad kinds: 1) open-ended, where you cannot be precise about what it is you need to know; and 2) close-ended, where you know exactly what you need to find out. As shown in Figure 6.10, open-ended data collection is more appropriate during the early stages of a project.

Open-Ended Data Specification

Here it is useful to create an exploratory agenda. Typically, it will consist of areas of interest of enquiry that you want to explore, with perhaps subsidiary topics that you want specifically to look at.

The specification can be generated simply from the sort of cause and effect analysis shown in Figure 6.5. The main legs of the diagram could provide the areas of enquiry, the branches from them could point towards the topics that you want information on. So, the data specification could appear as in Figure 6.11 below.

Area of enquiry

Topic

  1. Systems

1.1 Operative waiting time

1.2 Machine downtime

1.3 Wastage and rework rates

1.4 Method

  1. Equipment

2.1 Machine downtime

  1. Materials

3.1 Supplies prices

3.2 Stock levels

3.3 Customer expectations of quality

  1. People

4.1 Manning levels

4.2 Downtime costs


Figure 6.11: Open-ended data specification

In this example, taken from the International Cutlery Company case study, each of the topics relates to an issue that John Smith may want to explore in more detail.

What often happens is that information gathered in the early exploratory stages prompts new ideas of what data is important to collect and the data specification is modified accordingly.

Close-Ended Data Specification

When engaged in close-ended data collection, you can specify more precisely:

  • what data you need;

  • what form it might take - for example, what units it is measured in;

  • where you might find it.

Specifying precisely what data you need becomes particularly important when working in teams of consultants; each member of the team must be clear what data he or she has to collect. For example, a multi-country study would be of little use if the information from each country was not compatible, because different consultants had interpreted what was required in different ways.

To illustrate the process of data specification, we will again consider the ICC case study, but at a level 3 intervention - one at which the issues have been confirmed and the work now consists of identifying how best they might be addressed.

Suppose that the issue John Smith is to investigate is that of overtime costs, and that he has selected hypotheses for the problems underlying this issue as follows:

  • There is too much work for the labour force.

  • People are not working hard enough.

  • People are not sufficiently skilled.

  • Work is poorly planned.

  • Overtime pay rates are high.

Practice differs in consultancies in how they set out a data specification. Some consultancies identify what is called a 'key question related to each hypothesis'. This is a question, which if answered, would help you to judge whether the hypothesis is true.

The perfect question would simply be an inversion of the hypothesis: thus in the list of hypotheses quoted above, the hypothesis 'too much work for the labour force' would invite the key question 'Is there too much work for the labour force?' Indeed, some consultancies do not distinguish between hypotheses and key questions - confusingly, they call them all 'questions'. I prefer the split into hypotheses and key questions, as a key question can be phrased to illuminate one part of a hypothesis.

Key questions are particularly helpful if the hypothesis is expressed very generally and has to be focused before any work can be done in investigating whether it is true. This depends on the degree of abstraction of an idea. An example of increasing degrees of abstraction are as follows:

  • Spot is a small, playful, furry thing.

  • Spot is a puppy.

  • A puppy is a pet.

  • A pet is an animal.

'Spot' is a specific, concrete idea; 'animal' is far more general and abstract. A hypothesis about Spot will be easier to verify than one about animals in general. The 'key question' technique can therefore be used to help to reduce the degree of abstraction of a hypothesis. In the ICC case study, for example, you might come up with the hypothesis 'quality is poor'. This is abstract and difficult to verify, but could be made more concrete through key questions such as:

  • Are wastage rates above the industry average?

  • Do customers return goods as being of inadequate quality?

  • Are quality standards set too high?

Once the key question has been phrased, a data specification can be prepared. A data specification defines:

  • what data is required;

  • what format it might be in;

  • where it might be found.

The data specification also shows checks: where there is a particularly crucial piece of data that is required, you may wish to have more than one source of it. Figure 6.12 shows a completed data specification for collecting information about International Cutlery Company's overtime costs and how they compare with those elsewhere.

click to expand
Figure 6.12: Close-ended data specification

Compromise is Often Necessary

Data is expensive to collect, in terms of time, and thus in money. Often you have to determine what data is necessary and sufficient, and then possibly to compromise still further. The need for compromise is raised here, because hard data is sometimes difficult or impossible to collect. For example, it may be desirable to know the market prices of each supplier of widgets across the world, but official statistics may not cover this, and the manufacturers themselves could be reluctant to provide them to you. Alternatively, you could be subject to disinformation. A compromise may be made by substituting qualitative for quantitative data, and opinion or consensus in the absence of objective data.

The latter is particularly necessary with soft data - that relating, for example, to views and opinions. Soft data is important when dealing with recommendations or matters of implementation, when recognition of the political climate, and the culture and values of an organization, may be essential in achieving acceptance. The more senior your client, the more likely he or she is to be concerned about soft issues and data, such as those concerning competitiveness, communication and morale.

Choosing a Method of Data Collection

There are four generic methods of collecting data:

  • face to face from other people;

  • remotely from other people, by using questionnaires or similar documents;

  • looking at documents and records;

  • direct observation.

Each has its pros and cons, which are summarized in Figure 6.13.

Method of data collection

Pros

Cons

Interviews one to one

Personal contact with the interviewee

Unstructured - you can follow up points of interest

The interviewee has made a clear contribution

Enables you to judge what sort of person the interviewee is

Time-consuming

Difficult to decide who to see Time-consuming to analyse

Interviews one with a group

You can meet more people

The project has a higher profile

Hard work - probably needs two people

Less opportunity for individuals to contribute

People may be inhibited from contributing

Questionnaires

You can collect a large number of views

A well-designed questionnaire should be easy to analyse

The respondent can fill in as and when he or she wants

Is not time-consuming for client staff

Close-ended; you get answers only to the questions you ask

Must be self-explanatory

Respondents may have reservations about committing their views to writing

No sense of strength of feeling or relevance (although you can put in scales to test this)

Low response rates

Document inspection

Good chance of getting unexpected data

You can go at your own rate

Limited availability of documentation

Time-consuming for consultant

Can be difficult to find the data wanted

Observation

First-hand information

Good chance of picking up something unexpected

Observation can affect the system being observed

Time-consuming Difficult to analyse


Figure 6.13: Pros and cons of different methods of data collection

Pros and Cons of Different Methods of Data Collection

Quite apart from their intrinsic merits, when you have to choose a method for collecting the data you need, there are four other criteria to be met:

  1. Is it sufficiently open-ended? Will it collect the data required on the hypotheses being explored? It is important that wrong assumptions are not built in (for example, as in the question 'When did you stop beating your wife?')

  2. Will it collect the 'soft data' required? This relates to people's opinions. These will be particularly important in considering the acceptability (or otherwise) of recommendations, for example. Hard data leads to idealistic solutions; soft data provides information on how to make them workable and acceptable.

  3. What will its impact be? Remember that data collection is an intervention into an organization. This, if poorly handled, can have dysfunctional effects. On the other hand, the data collection method can be used to suggest processes of desirable change, or to give the project a suitable profile in the organization.

  4. Is it economical and effective?

Finally, remember that as data collection is itself an intervention into the client's organization, it is not possible for you to carry it out without in some way affecting the views of the client's staff about you, your practice and the project you are undertaking. A well-constructed and executed data collection plan can enhance your credibility, yield high-quality solutions and help to ensure more ready acceptance of your recommendations.

Interviewing Skills

Much of the data will be collected using interviews. Your interview plan should reflect the type of data that you require to answer the key questions related to your hypotheses. Bear in mind that you could have objectives for the meeting other than data collection, such as:

  • building relationships with the interviewee;

  • giving information;

  • canvassing support for a particular view or opinion;

  • problem solving;

  • decision making.

Remember to leave sufficient time in the programme for data analysis as well as collection; for example, to write up or consolidate interview notes and to relate the data to your hypotheses. Interviewing requires sustained concentration and is tiring, so do not try to pack in too many interviews in a short time. A good rule of thumb is to allocate twice as much time to an interview programme as the time you expect to spend face to face with interviewers. Try to structure data collection in a logical order so that key questions are answered first, and so that you avoid carrying out data collection that subsequently proves needless. Aim to get the data with the greatest 'leverage' first. For example, if the support of a Managing Director is crucial to the acceptance of your proposals, you need to find out if he or she has any strong dislikes or preferences fairly quickly, and this could condition which hypotheses you select to pursue.




The Top Consultant. Developing Your Skills for Greater Effectiveness
The Top Consultant: Developing your Skills for Greater Effectiveness
ISBN: 0749442530
EAN: 2147483647
Year: 2003
Pages: 89

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net