Measurement of Marketing Phenomenon Essay Example
Measurement of Marketing Phenomenon Essay Example

Measurement of Marketing Phenomenon Essay Example

Available Only on StudyHippo
  • Pages: 8 (2024 words)
  • Published: March 24, 2018
  • Type: Research Paper
View Entire Sample
Text preview

Different types of scales, including numeric scales, semantic scales, and graphical scales, are used in marketing research. It is crucial for marketing researchers to have a comprehensive understanding of all scaling measurements to gain insights into markets.

Marketing research texts typically outline the four levels of measurement: nominal, ordinal, interval, and ratio. Nominal scales are the simplest form of measurement scales where entities are categorized without any implied order. They are also known as categorical scales. The frequency of cases assigned to each category is counted and labels can be assigned to each category. An example of a nominal scale is shown in Figure 3.1:

Figure 3.1 An example of a nominal scale:
  • [ ] Milled Rice
  • [ ] Pastured Milk
  • [ ] Kara
  • [ ] Peppers
  • [ ] Palm Oil
  • [ ] Prawns

gn="justify">The numbers in the scale serve solely as labels and do not possess any arithmetic properties. The mode is the only measure of average that can be utilized since it represents the frequency counts. Hypothesis tests can be conducted on data collected in nominal form, with the Chi-square test being the most commonly used.

The Chi-square test is used to determine the association and strength of relationship between variables. Ordinal scales are discussed in the text, which are useful for ranking individuals, attitudes, or items based on a characteristic. While ordinal scales do not provide information about the nature of a relationship or establish cause and effect, they can still obtain all the information provided by nominal scales. Additionally, ordinal scales can be used to calculate positional statistics such as median, quartile, and percentile. The text mentions two methods for testing order correlation with ranked data: Superman's Ranke

View entire sample
Join StudyHippo to see entire essay

Correlation Coefficient and Sandal's Coefficient of Concordance. These methods assess agreement among survey respondents in their rankings of a set of items. An example shown in figure 3.2 illustrates the ranking of pesticide brands using this coefficient calculated based on respondents' farm enterprises classified as "arable" or "mixed". This coefficient ranges from 0 to 1, with 0 indicating no agreement between the two groups and 1 signifying complete agreement. Other acceptable hypothesis testing procedures include the runs test and sign test, where the runs test (also known as Wald test) determines if there are systematic runs of one value or if it is a random distribution.This test only considers two possible values (e.g., African/non-African, yes/no, male/female). Sign tests are used to compare matched pairs of data and determine if there are significant differences. They focus on the direction of differences rather than their magnitude, which makes them suitable for ordinal data. Interval scales allow researchers to use the arithmetic mean as an average measure. These scales have equal units of measurement called interval or cardinal scale scores that can be interpreted and compared. However, it is important to note that the zero point on an interval scale is arbitrary and does not represent a true zero. This characteristic affects how data collected in this form can be manipulated and analyzed. Adding or subtracting a constant value does not affect the scale on an interval level, but multiplying or dividing values violates its principles. For example, two respondents with positions 1 and 2 on the scale would be considered equally distant from each other as those with positions 4 and 5; however, it cannot be inferred

that someone with a score of 10 feels twice as strongly compared to someone with a score of 5.Temperature is measured using an interval scale such as Centigrade or Fahrenheit.The text discusses the incorrect assertion that a temperature of 0 degrees Celsius is twice as hot as -3 degrees Celsius due to the ratio not being 2:1. It also mentions interval scales, which can be numeric or semantic, and provides three examples of interval scales in both formats. One example involves rating Balkan Olives on a scale from 5 (Excellent) to 1 (Poor) based on twelve specified criteria. Another example involves using ticked responses to evaluate Balkan Olives as Very Good, Excellent, Good, Fair, or Poor. Interval scales are extensively covered in basic statistics texts and are commonly used for statistical analysis methods. For more advanced measurement purposes, ratio scales are employed, which have the properties of interval scales but also incorporate a fixed origin or zero point. Examples of measurements taken on a ratio scale include weights, lengths, and times. Ratio scales allow for comparisons between score differences and relative magnitudes of scores. The text further explores measurement scales within sociological research contexts alongside management research and marketing research fields.
The text discusses the limited use of interval level measurement in research fields and highlights that ratio scales can be used for all statistical operations. It differentiates between imperative and non-comparative scaling, where comparative scaling involves comparing brands or products while non-competitive scaling involves evaluating a single product independently. Comparative scales, such as paired comparisons, are utilized to determine factors influencing product demand. An example is given regarding farmers' negative response to an animal-drawn mould

board plough due to various factors such as inability to ridge, unsuitability for inter-cropping, high cost, perceived riskiness of new technology, and transportation difficulties. Identifying the most significant factor is crucial for understanding farmers' mindset so that addressing major factors can overcome obstacles to widespread adoption. The responsible organization now has the option to either abandon product redevelopment or completely redesign it. Both options - purchasing the plough or not purchasing it - are expensive and time-consuming and may lead to new objections. To prioritize these objections, a method called 'paired comparison' is employed. In this method, each factor is compared with every other one. The farmer is given one pair at a time to compare. To assess the importance of different factors leading to a decision against buying the plough, a question is posed to the farmer.The question is typically asked verbally as follows: "Which was more important in your decision not to purchase the plough - its high cost or difficulty in transportation?" The farmer determines the more important factor, and the researcher marks the corresponding checkbox on the questionnaire. This process is repeated with different sets of factors, marking checkboxes each time. All possible combinations are assessed, resulting in 10 pairs. It is recommended to randomize these pairs to avoid bias. The researcher should ensure that each factor appears first and second in comparisons at times, as comparing only the first factor with every other factor would be biased. Labels have been assigned to enhance understanding: A = Diodes not ridge, B = Afar too expensive, C = Nine technology too risky, D = Diodes not work for inter-cropping, E = Too

difficult to carry. The data collected from 200 farmers is organized into a matrix following a top-to-side reading pattern. For example, out of the 200 farmers surveyed, 64 indicated that the high cost of the plough was a greater deterrent than its incapability to ridge.Similarly, 174 farmers indicated that the plough's inability to inter-crop was more significant than its inability to ridge when determining their decision not to purchase it. Figure 3.4 exhibits a preference matrix with values assigned to different factors. Upon careful examination, it becomes clear that the most important factor is the crop, followed by the difficulty of carrying it. The matrix also suggests that designers should prioritize enhancing the plough's transportability and potentially incorporating an inter-cropping capability rather than focusing on its ridging capabilities. It is worth noting that this example is purely hypothetical.

One major advantage of this questionnaire type is that it allows researchers to determine the relative importance of multiple factors without overwhelming respondents. Respondents are only asked to evaluate two factors at a time. This approach proves particularly useful when dealing with illiterate farmers. However, caution must be exercised during the interview to prevent presenting too many factor pairs, as this can result in fatigue or disinterest from the farmer.

It should be mentioned that when there are ten factors or attributes, resulting in 45 possible pairs, one must avoid asking the same question 45 times to the farmer.The paired comparison scale provides ordinal data and has a limit of six factors, resulting in 15 pairs.The dollar metric comparison scale builds upon the paired comparison method by asking respondents to indicate their preference and the amount they are willing

to pay for it, thus providing interval-scaled measurement. An illustration of a dollar metric scale can be observed in Figure 3.5, where respondents express their preference for different types of fish and the additional amount they would be willing to pay for their preferred fish. This information can then be utilized to calculate preferences.

The unity-sum-gain technique aids in decision-making regarding the number and options to offer when introducing new products. While companies strive to cater to various market segments, they must ensure that the segment is substantial enough to justify investment. Expanding a product line is relatively easier, but determining which additions will succeed requires a unified approach. This involves creating a list of potential "options" for the product while noting their retail costs. A third column is created as an index of relative prices for each option (Figure X). The provided sample table demonstrates this process.

The base reaper has a cost of $20,000, and various extras along with their prices are listed. The cumulative value of these hypothetical additions amounts to $7,460. However, the researcher informs the farmer that only $3,950 is available within their budget. It is crucial for the farmer's available funds to be significantly lower than the total value of alternative features. This encourages farmers to express preferences and enables researchers to observe trade-offs between benefits.
The farmer's preference between a side rake attachment on a 3-meter head or a transporter trolley on either a standard or 2mm wide head should be based on finding the best value for their money. It is important for the farmer to consider this as they cannot keep any unused funds. To address concerns about bias

in results, an index can be used instead of specific prices. This index is calculated by multiplying each item's price out of a total of $7,460 by 100. Survey respondents are then given a maximum of 60 points and asked to allocate these points according to their preferences. Rounding has been applied in this example's adjusted column as using precise index numbers may be challenging for most respondents. However, precision in rounding is not critical as relative values matter more. The final market version of the product should be designed to meet farmers' needs and preferences. Data gathered through this method is treated as ordinal by practitioners. Noncompetitive scales, such as continuous rating scales where respondents mark their rating on a continuous line, are used during the interview and can be shown to the respondent on a card. Figure 3.7 displays two versions of the continuous rating scale which can have as many categories as desired. Respondents are assigned scores based on which category their mark falls into or by measuring the distance from either end of the scale in millimeters or inches. The results are typically analyzed as interval scaled regardless if categories or measurements were used.

Line marking scales are a commonly utilized method for measuring perceived differences in similarity between products, brands, or other objects. These scales fall under the category of semantic differential scales, where each end is associated with an opposite meaning word or phrase. An example of such a scale can be seen in Figure 3.8, which showcases how it determines similarities or differences in flavor among various types of frying food products.

Some individuals prefer the line scale format as

discrete numbers do not accurately represent their attitudes or emotions. The text highlights that there are two types of rating scales: itemized rating scales and semantic scales. Itemized rating scales have a limited number of categories with corresponding numbers or brief descriptions, allowing respondents to select the category that best reflects the product or attribute being studied. Figures 3.9 and 3.10 provide examples of these itemized rating scales, often treated as interval level data by researchers.

Semantic scales differ from itemized rating scales in that they employ words rather than numbers. Respondents use semantic labels to convey their sentiments about products or brands. If bipolar adjectives are utilized at the endpoints of these scales, they are referred to as semantic differential scales.

Get an explanation on any task
Get unstuck with the help of our AI assistant in seconds
New