How to evaluate site form filling efficiency use klm-goms to analyze form efficiency

Intermediary transaction SEO diagnosis Taobao guest Cloud host technology Hall

Editor: s++ Group

Many times it is difficult to test the efficiency of a form by user testing, and it is often the case that the form comes online and then counts the amount of time the user is using or staying to consider the efficiency of the form, but this can only be measured against a form that has been online. Is there any scientific way to evaluate the efficiency of a form in the design process?

1983 Card,moran & Newall First mentioned the Computer model in human-Computer Interaction psychology (the Psychology of Human Consortium GOMs). There was also a gust of wind in the field of early human interaction design.

So what is GOMs?

The so-called goms is Goals,operators,methods,selection rules, translation into Chinese is the goal, operation, method, selection rules.

Goals refers to the result of a task.

Operators refers to all the behaviors generated in the process, including control use, fill, select, click, suspend and so on operation.

Methods refers to the process of describing how to accomplish a goal.

Selection rules refer to the rule of judgment, which is the different rules followed under different conditions.

Why use GOMs?

GOMs can objectively record the whole process of how the user fills out the form. And it provides a scientific analysis method for the complex filling form process.

It provides a more efficient way of computing, in the process of testing different versions, can simply through him to compare the efficiency of the form design scheme.

By GOMs, you can obtain more reliable scientific data to prove the efficiency of the design of your form.

That GOMs has many branches? But what else can we use?

Cmn-goms,cpm-goms models, KLM-GOMS,NGOMSL and many more.

But in the form fill goms operation, we need to focus on the key to understand the keystroke model. It is referred to as KLM (Keystroke-level model). KML is a simplified model system, and he has relatively weakened the concept of goms to emphasize O, which is the operating part. 1988 Bonnie John Further proposed a simplified version of the cpm-goms theory.

The KLM model is a 11-step approach to finding the time to evaluate a task that completes a simple data entry using a computer and a mouse.

This model is used in many human-computer interaction areas, and he can accurately calculate the task time required for the overall form to fill out through task steps and operations.

Kieras (1993 2001) defines the following actions:

K, pressed key and release (keyboard) (wpm=words per minute)

Best input (135 wpm)-0.08 seconds

Good input (90WPM)-0.12 seconds

Poor input (40WPM)-0.28 seconds

Average skilled input (WPM)-0.20 seconds

Average non-secretarial input (40WPM)-0.28 seconds

Enter the letter at random-0.28 seconds

Complex code input-0.50 seconds

Worst input (different keyboard input mode)-1.2 seconds

P, pointing the mouse at the object on the screen-1.1 seconds

B, Button press or release (mouse)-0.10 seconds

BB, double Mouse-2 sec

H, the hand from the keyboard, mouse away or put on-0.40 seconds

M, mental preparation-1.2 seconds

T (n), String type character input (n * K sec.)

W (t), waiting for system users to respond

D (n0,l0), using the mouse to draw a line through the formula to calculate the time required

So what are the 11 steps in the KLM analysis?

Step 1--to obtain step-by-step instructions for prototyping or tasks.

Step 2--Find the desired final goal or desired work outcome.

Step 3--Find the task flow to achieve the primary goal.

The steps 4--the primary goal identified and the task flow in all subordinate targets.

Step 5--into code.

Step 6--uses code to describe the entire task flow process.

Step 7--First Keep each step of the psychology or operation.

Step 8--for each step of the distribution of psychological or operational, that is, the appropriate part of the unnecessary psychological time to delete.

Step 9--perform the operation.

Step 10--to adjust the total time of the task according to different age levels

Step 11--Validation Results

But the use of M (mental preparation) is the key. The placement of M (mental preparation) is even more important than the precise calculation of other operations. So make sure you use the same m placement rules when you're working with different design versions.

So for how to place the processing m (mental preparation), we have the following suggestions:

Start the task and the user has to stop to know what to do with the task and what to do with the task.

If there are many ways to make a decision, the user (who is not very familiar with the task or the decision is not obvious) often stops thinking before making the decision.

Need to get information from memory, such as recall file name, command, or abbreviation, need to have some time to think.

Looking for information on the screen, users have to stop and browse the screen to find out what information they can't predict through experience.

Considering the task reference value, the user sometimes remembers that the parameter is sometimes read, and in this case it is necessary to consider whether to increase m (mental preparation).

The right error of the checksum behavior, the user usually stops and checks their information or checks their entry before submitting the system response.

The placement of M (psychologically prepared) in the Web page will have some general principles:

K (keystroke keyboard), P (move mouse to target), B (click the mouse), M (psychological preparation), W (Waiting for system response).

Principle 1 Enter the content (e.g. all k, P, B) at the beginning and join M.

For example, if you move the mouse to a target, and then you click on the target (press the key on the mouse), or adjust the target, then according to principle 1 your time calculation is, MPMB;

Principle 2 Removes the expected m if the action from the back of M can completely infer the action before m, then remove the M.

For example, when browsing the picture, click on the picture to browse the next picture, and as a long time users have been fully aware of the operation will produce results, this time the calculation becomes, MPB;

Principle 3 Removes a series of m in the same behavior, removing all but the first M.

For example, input "Soso" four letters, according to the principle of 1,mkmkmkmk, according to the principles of MKKKK;

Principle 4 Removes the command non-terminal before the M. For those command characters with an action class followed by a long string of strings and habitually used for a long time, you can delete m.

For example, when landing a form, and very familiar with the time, in completing the user name, with the tab switch to continue to enter the password, we can delete M.

For example, when a command character is entered and an effect is known to the command, the m before the carriage return can be ignored. But there is a situation click on the button will not be able to predict the effect of the click, or the click of the button will virtually create psychological pressure behavior, this time m still need to be added back.

For example, when the purchase button appears, many users will have psychological hesitation, such as Delete button is the same;

Principle 5 Remove Duplicate M

When W, which is the system response time, the M needs to be deleted. #p # subtitle #e#

Let's take an example.

Let's compare the controls for A and B as a form, and the time it takes to change the options!

First, we need to break down the steps.

A hand on the mouse refers to the desired selection of the radiobox〉 mouse click the option.

b Put your hands on the mouse. Click on the dropdown box to select the option to click on the option in the Drop-down box you want.

And then we'll turn this into code.

Finally, calculate the time.

A, H (0.4) +m (1.2) p (1.10) +b (0.1) =2.8

B, H (0.4) +m (1.2) p (1.10) +b (0.1) +m (1.2) p (1.10) +b (0.1) =5.2

So according to this calculation the Radiobox control is in a sense more efficient than a drop-down box control.

Let me give you another example.

As shown in the figure above: There are three different designs for the translator, so let's take a look at him.

First, let's break down one step at a start.

A, put your hands on the mouse. "Point to the text box" Click on the text box "hand placed to the keyboard" input good〉 again put the hand on the mouse to move the mouse to the button OK click button

A1, hand put on the mouse to Radiobox English "click Radiobox English" refers to the text box "Click the text box" hand placed to the keyboard "input good〉 again put the hand on the mouse" move the mouse pointer to the button OK click button

B, hands on the mouse "point to the text box" Click the text box to put the hand into the keyboard to enter a

C, put your hand to the keyboard and enter OK (because the cursor is directly in the text box)

And then we'll convert to code.

Finally, calculate the next time.

A, H (0.4) +m (1.2) p (1.10) +b (0.1) panes (0.4) +m (1.2) +k (0.2) *4+m (1.2) panes (0.4) +m (1.2) p (1.10) +b (0.1) =9.2

A1, H (0.4) +m (1.2) p (1.10) +b (0.1) +m (1.2) p (1.10) +b (0.1) (panes) 0.4 () +m (1.2) (+K) 0.2 (*4+m) 1.2) panes (0.4) +m (1.2) p (a) ( 0.1) =11.6

B, H (0.4) +m (1.2) p (1.10) +b (0.1) panes (0.4) +m (1.2) +k (0.2) *4=5.2

C, H (0.4) +m (1.2) +k (0.2) *4=2.4

Then after analysis a choose English and Chinese, the probability of 50%, so a finally counted down is (9.2+11.6)/2=10.4

At a glance the most efficient translator here is C.

But GOMs is not a panacea. In the task flow, the rationality of the task itself can not be realized by goms, and these subjective analysis needs to be validated by a series of uer studies.

GOMs is difficult to analyze the user's understanding process, and can not analyze the interface of the "feeling" for the page in the Chinese case, including visual style caused by the user perception of cognitive goms completely unable to give any advice.

Therefore, although GOMS has its scientific basis and the existence of objectivity, it also has its unpredictable defects. He can only be used for the design of the task process time scientific calculations, but can not give visual and cognitive advice.

Therefore, we can not blindly pursue the efficiency of form filling, or should consider some of the control of the rationality or habitual use of habits, as far as possible to reduce the user in the form fill in the doubt, hesitation, irritability, on this basis to choose the most efficient form to fill out the best way to perfect the design.

This article author: s++ reprint please specify from: Ctrip ued