Gestalt Principles and Affordances on WSJ.com

Gestalt in the News

Analysis of rules seen in the front-page of www.wsj.com (The Wall Street Journal)

Rule #1 – Headings
Consistent formatting helps show what kind of text something is – tab, article title, body, or subsection header. The tabs are indicated by sans serif font (Whitney SSM or Arial), a slightly heavier weight font, small size, black. The titles of articles are indicated by serif font (Chronic Display or Georgia), bold, larger size, black. The body text of articles is serif font (Chronicle SSM or Georgia), regular font weight, small size, black. The subsection/topic headers are sans serif font (Whitney SSM or Helvetica), small caps, of various colors, but all colors have similar value.

Rule #2 – Borders
Borders divide up the page, and the border thickness and color denote how “deep” this divide is. Between completely different sections such as between the tabs and the content, there is a strong black thick border. Between different articles in the same section, there is a pale thin gray border.

Rule #3 – “Read More” Arrow
There is a “right arrow” for each section to read more. This rule is somewhat inconsistently applied. For the “Opinion” section this arrow is at the top to the right of the section title, and is close enough to the section title to be associated with the section title. The “Markets” section has its arrow at the bottom with an additional smaller “MARKETS” label. The “What’s News” has it at the top, but this header bar is so wide that the arrow all the way to the right is so far away from the “What’s News” at the left that in my mind they are not clearly associated.

Rule #4 – Data Visualizations
Data visualizations use a consistent shade of blue for the graphs. The text is gray sans serif.

Rule #5 – 12 Column Grid
The page appears to be laid out on a 12 column grid. In its widest layout, the four columns on the page are, from left to right, width 3, width 4, width 2, width 3.

Rule #6 – Image Aspect Ratio
The image aspect ratio is usually the same for images. This helps gives rhythm rather than chaos to the webpage, which is crowded with text and images.

*Gestalt Principles in play*

Contrast:
The WSJ website uses a reverse-contrast to distinguish between static content on the site (black foreground on white background) and the dynamic content represented by the stock ticker at the top (white foreground on black background). This according to us is a very good use of contrast as it in terms of scale (thin strip compared to the rest of the site), it creates a different context of lower relative emphasis.

Emphasis:

There are a few different ways the WSJ hompage communicates emphasis. The first one is through space. Note that the headings for each subsection of the paper are not much bigger than the rest of the text on the page. The way they stand out is that their entire right side is cleared, giving that word notable prominence, as nothing is on its same level, vertically, until you get to the next section. Another (more obvious) way is font size, as demonstrated by the top left headline. A third way is the use of center alignment for Opinion articles. Everything else is right-aligned, and the disruption of the rhythm created by this alignment is what sets off the section as important.

Balance:

We noticed some imbalance in WSJ.com’s header. The designers seem to have placed all the “extras” in the header on the right side, with nothing on the left to balance it. Subscribe and Sign In links are there, along with a red 50% OFF indicator.

Harmony:

All of the images in WSJ’s layout seem to be of the same 1.5 aspect ratio, despite their varying sizes. This adds a sense of consistency throughout. Also, they tend to use a consistent accent color (a mid-blue) seen in mouse-over transitions and in the Market section’s line graph.

Rhythm:

As I mentioned above, left justification is used almost exclusively throughout. There also seems to be consistent horizontal space between columns and consistent vertical space between vertical elements. As the eye travels down the page, one can see which articles are related through alignment of photos and text, such as the strategy employed with stacked boxes in the business section containing right-aligned photos of the same size surrounded by left-aligned text

Other Case Studies

In getting away from traditional new outlets, we wanted to consider NPR.org, whose news is generally disseminated in audio format, with text being secondary. This website is much simpler than WSJ.com, and after a menu on top, takes advantage of a pattern called “cards” to create a nice sense of rhythm. Each card is the same width with the same amount of space between itself and the next one. Each card is labeled with a category, and most feature a headline, a picture, and a button that allows the user to listen to a story. The type treatment is consistent across roles, so without reading the text, you know what information it’s going to tell you.

Another news website that our group studied which is quite unconventional in terms of visual representation was Slate (www.slate.com). Slate again used reverse contrast but this time, augmented it with scale to add emphasis to the most important article on the front page. To indicate difference in context of the link to the Facebook page as well as the request to disable the Ad Blocker (both of which are lower in hierarchy compared to the main news content), Slate glued thin strips of reverse contrast to the top and bottom of the page. As you scroll down further, there is a clear shift in weight to the top right as the heavy Slate logo stays glued to the top right of the page. This, we thought was an excellent manipulation of balance as users can find the menu and the search option on this ‘heavy’ block.

Al Jazeera, as we studied also had some interesting variations on contrast, emphasis and overall balance across its American, English and Arabic versions. The American version had a visual identity that was quite distant from its English and Arabic counterparts and used scale to indicate hierarchical difference in menu, space to indicate emphasis of the titles and contrast to indicate selected menu item. The English and Arabic version, while more similar at a glance, had some differences in the way they used balance and contrast. The English website was a more Left-to-Right oriented website with the emphatic elements predominantly on the left while the Arabic version followed the opposite trend. However, while the live news update was reverse contrasted, it stayed on the left side of the screen for both versions.

The Guatemalan newspaper Prensa Libre appears well-balanced at the top of the web page, but the design rules do not stay consistent as one scrolls down. In the header, the wide white side bars, use of free space and symmetrical layout create a feeling of balance. The distinctive brush-like font used for “Prensa Libre” contrasts dramatically with the Palatino/ Arial font of the rest of the page. The consistency of the font types, blue/gray colors and balanced arrangement create a sense of harmony. The main content is arranged in three columns of unequal width, with some emphasis given to the headline photo on the upper-left that takes up two columns, and greater emphasis given to the column on the right which is slightly wider than the rest and demarcated with a red line and small red clock symbols highlighting the latest news updates. Since the rest of the text and formatting lines are blue, black or gray, the red contrasts dramatically. The red clock symbols match the one in the drop-down menu, creating a clear consistency. On the other side of the page, the upper-left corner is grounded by a large black box with a simple white outline of a house; our eye is clearly drawn to the homepage link. As we scroll down the page, we see a balanced rhythm of column layout alternating image/ text/ image / text. However, as we scroll further down the page, we hit a shadowed line break where the “headlining news” ends and the “Mas noticias” and “Editorial” sections begin. The right column drops out and the other columns do not realign; instead, we are left with a glaring white space which feels unbalanced. For the first time, we have a left-aligned column of bullet points. The inconsistency is startling – I suppose that it is effective in making us feel like we have moved into another section of the paper.. At the bottom of the page, we hit another shadowed line break and the layout shifts entirely for the “Opinion” section; now we have four equal columns with titles that are center-aligned. If we were to study the “Opinion” section in isolation it would appear well-balanced, but the dramatic shifts in layout between sections destroys the overall harmonicity of the page.

Data Visualization Workshop: Are LA Police Stops Effective?

TEAM #2 City/Mayor

CONTEXT: This is an internal, actionable report on the effectiveness of police stops for the Los Angeles Mayor. Our analysis looks at the police stop accuracy rate, i.e., the percentage of stops that are immediately followed with post-stop activity (as opposed to unnecessary stops which do not justify further activity). We showcase the 5 most effective and 5 least effective neighborhoods, and then further breakdown the police stop effectiveness by race for each neighborhood.

ACTIONABILITY: This data could be used to identify the top performing police precincts, reward their success, and analyze what specifically is superior in their policies and practices. Ultimately, the goal would be to transfer those policies, practices and perhaps even personnel to the lowest performing precincts.

What is your story?

Working in the mayor’s office, we have decided to create an internal, actionable report for the Mayor that cuts to the essentials of what can be quickly acted upon due to the Mayor’s limited time. We focus on the ratio of “Stops with Post-Stop Activity” / “Total Frisk Stops” in ten different neighborhoods across the city to convey the efficacy of Stop-and-Frisk legislation. We focus on the top five and bottom five precincts in order to best understand which practices lead to their success or failure. Additionally, we break down the police stop efficacy by race for each neighborhood to explore the possibilities of racial discrimination in the law’s application.

How does the selected data support your story?

The data allows us to ask, “Do some people get stopped and frisked for no reason?” The answer is yes, the residents of North Hollywood, West Valley, West LA, Devonshire and Pacific are over twice as likely to be stopped for no reason as the more effective neighborhoods like Foothill, Seventy-Seven, Central Bureau, Mission and Newton.

The data allows us to ask further, “Is there racial bias in who is stopped for no reason?” Given the national statistics in white versus black treatment by the police, one might assume that blacks are more likely to be stopped for no reason, but surprisingly, this “police stop efficacy” measurement is roughly equivalent between whites and blacks in all the neighborhoods. In other words, when blacks and whites get stopped and frisked in Los Angeles, they have a similar likelihood of experiencing further police activity, which suggests that blacks are not suffering discrimination. One surprising statistic is that Asians are the most likely to be stopped and frisked for no reason; in every single neighborhood, the “police stop effectiveness” is the lowest for Asians than for any other race.

Despite these interesting trends, our data is missing several key metrics: are more blacks getting stopped overall? We do not look at absolute numbers. Are fewer Asians stopped on average? We do not take into account the population balance of the different races in each neighborhood. It is possible that blacks underrepresented in the overall population but overrepresented in the percentage of stop-and-frisk incidents. Also, if blacks have an equal or even slightly higher percentage of post stop activity, is that really because police “accuracy” is higher, or could it be because whites get let off the hook slightly more often? Our chosen metric of “stop accuracy” uncovers surprising insights, but begs further investigation.

What data did you omit and why?

The most significant omission was the data (race and ID number) about the particular officers that were carrying out this law. Though these could be important for discovering the “accuracy” (in terms of the aforementioned ratio) of various officers as well as their personal, racial biases, its is both out of the scope of our report and outside the purview of the mayor to scrutinize individual officers in a city whose police force numbers over 10,000. Following the same logic of including only the information that could be absorbed in a limited window of time, other variables such as time of day, date of stop, and stop type (vehicle or pedestrian) were omitted.

How does the representation support your story?

The representation supports our story because bar graphs are simple to interpret, and keeping the y-axis the same for all graphs makes it easy to make comparisons between different towns and different races. For example, bar graphs made it easy to see that blacks and whites have about the same “stop accuracy” with it being slightly higher for blacks. Simplifying the color of the background and bar graphs makes it instantly recognizable which towns have more effective police stop policies (blue) versus those that do not (orange).

What visual metaphor(s) did you use and why?

The bar graph was chosen because it was the most immediately comprehensible. Though we considered using a map with pie charts and various layers with more sophisticated graphics, owing to the limited amount of time provided by our scenario of meeting with the mayor, we agreed that bar graphs would be the most efficient means of communicating the necessary information. The more effective neighborhoods are coded blue to signify well-being and stability, whereas the least effective neighborhoods are coded orange to signify danger and instability.