Let's say you wanted to investigate the possible relation between perfectionism and fearfulness. If you wanted, you could go to the IPIP home page, follow the link to the list of over 3,000 items, and build your own perfectionism and fearfulness scales from items that look like they measure those two constructs.
You would probably be better off, however, looking for existing IPIP scales for those constructs. The advantage of using existing IPIP scales is that they are known to be reliable and to correlate significantly with established, valid scales. To see whether IPIP scales for constructs of interest already exist, browse to the IPIP home page, follow the link labeled "Index of Individual IPIP Scales," and search for perfectionism and fearfulness.
A search for perfectionism will bring you to a line that reads
Perfectionism (AB5C: III+/IV-), (CAT-PD: Perfectionism), (CHS: Foa, et al., 1998), (HEX: C-Perf).
Each one of the abbreviations in parentheses is the name of an IPIP scale that seems to measure perfectionism. These scale names are also hyperlinks, so if you click on them you will be taken to a page that contains the items. You can examine the content of the items for these four scales and choose the scale that you think best suits your research purposes. (You can of course use more than one perfectionism scale if you think that would be useful.)
A search for fearfulness brings up the line
Fearfulness (HEX: E-Fear).
In this case, it seem that you have only one choice. But if you are unsatisfied with the content of this fearfulness scale, you can think of a word whose meaning is similar to fearfulness, for example, anxiety. A search for anxiety reveals five scales that might possibly suit your research purposes. Another way to find additional scales for your construct of interest is to think of the opposite of the construct. Courage could be considered the opposite of fearfulness, and a search for courage reveals one IPIP scale for measuring that construct. If you score the items on this scale in reverse, the score would represent lack of courage, or fearfulness.
If you are interested in measuring a broad range of personality constructs instead of one or two individual constructs, browse to the IPIP home page and follow the link labeled "Index of Multi-Scale IPIP Inventories (including measures of the five major personality factors)." This will bring you to a page of links to IPIP inventories designed to measure constructs similar to many well-validated, multi-scale inventories, including a number of measures of the five major personality factors.
Currently one of the most widely-used approaches to measuring the full range of normal personality is to use an inventory based on the five major domains of personality revealed by factor analytic research. These five domains are often referred to as "The Big 5" (B5) or "The Five-Factor Model" (FFM). While many nonprofessionals outside the field of personality psychology or newcomers to the field have heard of the five major personality factors, not everyone realizes that the B5 and FFM derive from two historically separate research programs (although Raymond Cattell played a role in both programs) and are based on entirely different kinds of data. The five factors were first identified from factor analyses of individual trait words (such as such as talkative, goodnatured, responsible, calm, and imaginative) found in the dictionary. Since the trait words came from our ordinary language (lexicon), this program of research is often called the lexical research tradition. Later, researchers who were aware of the five factors identified by lexical research decided to construct personality questionnaires based on those five factors.
Lexical research began with Gordon Allport's cataloging of roughly 18,000 trait words from an unabridged dictionary. Next, Raymond Cattell reduced the list to a smaller set of terms by various means including factor analyses of ratings of individuals. Later, two Air Force researchers, Tupes and Christal, reanalyzed Cattell's data and analyzed some new data with different methods, and found five factors every time. Their findings appeared in an obscure technical report, but were noticed by Warren Norman at the University of Michigan, who replicated their results and published them. He labeled the five factors Extroversion/Surgency, Agreeableness, Conscientiousness, Emotional Stability, and Culture. Norman's colleague, Lew Goldberg, continued to replicate his findings with different methods and sets of trait words, always finding the same factors, although he labeled the fifth factor Intellect instead of Culture. Goldberg is credited with coining the term "the Big 5" for the recurrent five lexical personality factors.
Personality psychologists have constructed a number rating forms containing different sets of personality trait adjectives, and the IPIP website contains IPIP scales that correlate strongly with these adjective rating forms. All of these measures are listed at the top of the Multi-Scale Inventory page under the heading Lexical Big Five Inventories. These inventories include IPIP versions of Goldberg's (1992) Big-Five Factor Markers, Saucier's (1997) 7 Factor Scales, 45 facets of the Abridged Five-Factor Circumplex (AB5C; Hofstee, De Raad and Goldberg, 1992), 10 Aspect of the Big 5 (DeYoung, Quilty, & Peterson, 2007), and the Interpersonal Circumplex (Markey & Markey, 2009). Some of these IPIP scales have shorter and longer versions; for example there are 50-item and a 100-item IPIP versions of Goldberg's (1992) Big-Five Factor Markers.
The above paragraph, which describes only IPIP scales based on trait adjective ratings, should make it clear that there is not one "Big Five Inventory," but many different IPIP inventories for measuring the B5 factors. The paragraph below describes additional IPIP inventories for assessing the FFM factors. It is up to you, the user, to decide which inventory best suits your research purposes. We should also note, in passing, that the expression "Big Five Inventory" (BFI) is the name of a 44-item personality inventory constructed by Oliver John. The items on the BFI are, like IPIP items, brief phrases rather than adjectives, but the BFI does not use IPIP items. There is no IPIP version of the BFI. If you are interested in the BFI, browse to Oliver John's website.
The second history of the five personality factors is based on personality questionnaires. Raymond Cattell had constructed a questionnaire, the 16PF, which contained 16 different scales. Paul T. Costa, Jr. and Robert R. McCrae, who conducted major longitudinal studies, first in Boston and then in Baltimore, used the 16PF, but believed that the content was highly redundant. So they analyzed the items on the 16PF and found three factors they labeled Neuroticism, Extraversion, and Openness to Experience. Eventually they constructed their own original personality inventory to measure these three domains of personality, and the called it the NEO Personality Inventory (NEO PI; Costa & McCrae, 1992). Later, after they heard a talk about the Big 5 by Jack Digman, a colleague of Lew Goldberg, they added an Agreeableness and Conscientiousness scale and renamed their instrument the NEO PI-R (R is for Revised). Costa and McCrae then proceeded to correlate their NEO PI-R with other major personality questionnaires and demonstrated that essentially all of the scales in existing personality questionnaires are related to the five factors in the NEO PI-R. Their view, which they call the Five Factor Model (FFM), is that Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to Experience are the five basic factors that underlie personality. The differences between the Big 5 factors and the factors of the FFM are not very large. The first four factors are nearly identical (Emotional Stability is simply the opposite of Neuroticism). The greatest difference lies in the fifth factor, where Intellect measures a tendency toward intelligence and an intellectual style, while Openness to Experience measures creativity, imagination, and an interest in trying new things. But the items on the IPIP Intellect and Openness scales possess similar content.
The IPIP scales that represent the NEO PI-R were created by identifying items the correlate highly with Costa and McCrae's NEO PI-R. The Multi-Scale Inventory page contains several IPIP versions of the NEO PI-R. Included are 50-item and 100-item measures of just the five factors. These measures are labeled "NEO PI-R Domains." Costa and McCrae's NEO PI-R also includes six "facet" scales for each of the five NEO domains. The IPIP scales associated with the label "NEO PI-R Facets" were developed to measure constructs similar to these NEO facet scales. There are currently three IPIP inventories that measure NEO facets: a 300-item version (Goldberg, 1999) that is one of the first IPIP measures ever constructed, a 120-item subset of the 300-item inventory developed by Johnson (2014) with internal consistency methods, and a 120-item subset developed by Maples, Guan, Carter, and Miller (2014) with item response theory.
The NEO PI-R is not the only personality questionnaire that was explicitly informed by the earlier lexical research. Long before Costa and McCrae added faceted Agreeableness and Conscientiousness scales to their original NEO PI, Hogan (1986) designed his Hogan Personality Inventory (HPI) around the Big 5. The 1986 version of the HPI contained six domain scales. Sociability and Ambition represented different versions of Extraversion; while Likeability, Prudence, Adjustment, and Intellectance corresponded to Agreeableness, Conscientiousness, Emotional Stability, and Intellect in the Big 5 model. A revision of the HPI (Hogan & Hogan, 1992) later divided the fifth domain scale into School Success, which retains closeness to the Big 5 Intellect factor, and Intellectance, which resembles Costa and McCrae's Openness to Experience construct. Domain scales of both versions of the HPI are divided into smaller, narrower scales scales called homogeneous item composites (HICs). The original HPI contained 43 HICs; the 1992 version revised the item composition of HICs and altered the scale assignment of these 44 somewhat different, new HICs. The IPIP site contains scales representing the 1992 version of the HPI.
The Multi-Scale Inventory page also lists IPIP versions of two more recent inventories that were explicitly based on the five major personality factors. The Six Factor Personality Questionnaire (6FPQ; Jackson, Ashton, & Tomes, 1996) contains domain scales for Extroversion, Agreeableness, and Intellectual Openness that correspond to the three factors with similar names in the FFM. The 6FPQ breaks Conscientiousness into separate Methodicalness and Industriousness domain scales. Independence corresponds to low Neuroticism. Each of these domain scales contain three narrower facet scales. The IPIP version of the 6FPQ can be scored for the six domains and 18 facets. The other FFM-based inventory is the HEXACO-PI (Lee & Ashton, 2004). The HEXACO-PI domain scales Extraversion, Conscientiousness, and and Openness to Experience correspond to the factors measured by the NEO PI-R. The HEXACO-PI Agreeableness and Emotionality scales are rotational variants of the NEO PI Agreeableness and Neuroticism constructs. Finally, Honesty-Humility is a sixth factor beyond the traditional five factors.
The Multi-Scale Inventory page also contains links to IPIP inventories designed to measure constructs similar to many widely-used inventories that are not based on the five major personality factors. These include the 16PF, California Psychological Inventory, Temperament and Character Inventory, Multidimensional Personality Questionnaire, and the Jackson Personality Inventory.
If you would like to measure constructs from more than one of the above questionnaires while keeping the total number of items relatively small, Yarkoni's (2010) Analog to Multiple Broadband Inventory (AMBI) can measure 203 constructs similar to those measured by the scales of the NEO PI-R, HEXACO PI, TCI, HPI, 6FPQ, CPI, JPI-R, and MPQ with only 181 items.
Most multi-scale IPIP inventories were constructed to assess constructs similar to those on established personality inventories by locating items that correlated empirically with scores on those measures. However, some multi-scale IPIP inventories were constructed to assess theoretical constructs of interest by locating items whose content seemed to tap those constructs. These inventories are listed toward the bottom of the Multi-Scale Inventory page.
For her doctoral dissertation, Kim Barchard (2001) administered seven purported self-report measures of emotional or social intelligence to the Eugene-Springfield Community Sample. The Multi-Scale Inventory page links to seven IPIP scales representing components of self-reported emotional or social intelligence. Barchard cautions that these self-report measures do not correlate with maximum-performance tests of intelligence and should be regarded as measures of personality rather than cognitive ability.
Other new multi-scale IPIP inventories include Peterson and Seligman's Values in Action Survey of Character (Diamond, O'Brien-Malone, & Woodworth, 2010), Simms, et al.' (2011) Personality Disorder Scales, Pozzebon, et al.'s Oregon Vocational Interest Scales (ORVIS), and Goldberg's (2010) Oregon Avocational Interest Scales (ORAIS)