5 användare borde vara tillräckligt för alla

Den där grafen ni vet, som Jakob Nielsen kläckte ur sig runt 1993, om att 5 användare är tillräckligt många att testa med. Den stämmer ganska bra, tycker jag. :)

fem

Jag börjar få ganska många års erfarenhet av användningstestning och om målgruppen är hyggligt homogen så börjar jag efter 3 test se samma problem dyka upp. När jag är uppe i 6-7 test så är det helt klart så att jag har fått mycket bra koll på läget. Det blir också tydligt då om målgruppen faktiskt är heterogen och vilka undergrupperingar det rör sig om. Efter 4-5 möten med användare i en subgrupp så vet jag om den gruppen är homogen.

Det är faktiskt ganska coolt.

Användarstudier med barn

Jag har inte haft en blogg sedan urminnestider, närmare bestämt sedan september 2008. I den här bloggen kommer jag att behandla agilt och ux, både högt och lågt. Det blir nog mest lågt, korta kommentarer, länkar och tips. Men jag tänkte börja högt, genom att återpublicera det sista blogginlägget i förra bloggen. Då körde jag på engelska, så ni får stå ut. Det här handlar om lillgamla barn.

Martin


A common way of separating users with the intent of forming different target groups for usability testing is to use age as a variable. The problem is that this is of course a tricky variable to use. Usually, we aim to define end-user populations as homogeneously as possible, since it is then easier to do user testing and optimize user interface design. This approach is good as long as the product is intended for professional use, by users who are between 18 and 50 years old. This approach was good enough twenty years ago when most computer programs had that target group. Today the situation has changed. Target user populations have become heterogeneous, with people of all ages using computers more or less on a daily basis and more or less successfully. Two outliers in this distribution are children and the elderly. Digital natives are a large user group and the computer literates has become twenty years older as well.

During these two periods, adolescence and senescence, the changes on a cognitive level and the differences between individuals are great. The former period we call development and the latter, for lack of a better word, decline. For example, you have probably heard about both precocious kids and late bloomers. A uniform age group is not applicable, hence it is important to have usability professionals taking a deeper look at user group compositions when developing for the youngsters and elder people of today.

I once carried out several projects involving both these outliers and became very intrigued by developmental psychology, especially the parts focusing on children. This article will focus on the theories behind and will try to give you some heads-up before you conduct usability testing with children.


© MrsGooding

One theory of cognitive development of children, which seems to be the most common, is the one of Jean Piaget. He divides childhood into four development stages; sensori-motor period (infants), preoperational period (2-7 years), concrete operational stage (7-11 years) and formal operational stage (11 years and up). Although the timing may vary, the sequence of the stages does not, thus this theory is a good base for differentiating target groups of children. The characteristics of the preoperational period is a development of an inner representation of external objects. The thought processes of the child are then linked to the most prominent features of an object or a situation. The concrete operational stage is characterised by logical and purposive thinking, although the operations are always connected to the actual situation. In the formal operational period the children disengage from the concrete situation and become able to perform systematic analysis on an abstract level.

Apart from standard interview techniques such as commencing the interview with smalltalk, there are some issues to take into consideration when interviewing children that I learned during the aforementioned projects. One such thing is considering the attention span of the child. A session with young children demands a flexible setup of the evaluation. The child should be able to explore the product almost on their own instead of following a set of tasks. They are often motivated by making adults happy, hence let them show what they have found in the product and increase their motivation by encouraging them. For example, say ”Wow, did you do all that by yourself?” or ”Is that how it works! Thank you for telling me!”. Furthermore, avoid placing the child in front of the interviewer, place the child in front of the product with the interviewer acting as support on the side. In one of the projects, four male interviewers in their early twenties sat down in front of an eight-year-old girl and were surprised when she didn’t want to cooperate in the evaluation. Apart from steering clear of these kinds of situations, it is a good idea to have younger children evaluating in pairs where they can encourage each other and share ideas. It is also easier for them to speak about the product with a peer than with an adult.

This shyness towards adults is most visible in the preoperational stage. Hence, children in this stage can also have problems with expressing their feelings for the product in words, especially in front of a grown-up. Observe their behaviour, sighs, smiles or if they simply disappear under the table (which occured a lot with some children). Also, try to avoid asking the children if they wish to play a game or perform a task as this will give them the option to say no. Instead, say ”Now, I would want you to…” or ”It is time that we…”. This is easier to do with children in the concrete operational stage.

Children in the concrete operational stage have a high tolerance for complex interfaces. They employ pattern-based problem solving, ”push twice on the left button and three times on the right button to reach the gold” comes natural as long as it benefits them. They are starting to understand how to critically review the task given to them. They will be able to answer questions regarding the task and try new approaches with joy, but they are very aware that they are being observed. The previously mentioned eight-year-old asked the interviewers, in a latter session, why they wanted her and not her sister to evaluate the product. When the session ended it seemed as if she only criticised the method and did not care about the product. Later on, her teacher collected some drawings of hers containing references to the product and it was accompanied with a sun and a couple of green trees. Some children would like to answer questions orally and other in writing, but remember not to neglect those who want to express themselves with pictures.

In the formal operational stage, children might be able to think aloud while they are performing a task. However, what has to be taken into consideration is that these pre-teens or teens are not geniuses that can adapt to every complex situation. The possibly bad performance of the teens are caused by mainly three factors; inadequate ability to read, bad research strategies and a relatively low level of patience. There are simply a lot of other things happening in their world at that particular moment that we unfortunately have to contemplate.

On the other hand, most children tend to be smarter than you would give them credit for. One eleven-year-old girl demonstrated her own Klik & Play-made programs to the interviewer and explained how she could make the product in question smarter. The outcome was, needless to say, very appreciated by both parties. Children understand the concept of usability. Most children in the two operational stages can spot the difference between fun and efficient. They are as motivated by reaching their goals as adults are, and they really do not like when a product is not working.


I början av 2014 höll jag en presentation om mina erfarenheter med användarstudier med barn. Här är mina slides, håll tillgodo.